sneak/routewatch
1
0
Fork 0
Code Issues 12 Pull Requests 5 Actions 50 Packages Projects Releases Wiki Activity

Compare commits

merge into: sneak/routewatch:issue-27-cached-stats
Branches Tags
sneak/routewatch:main sneak/routewatch:issue-52-revendor-prompts sneak/routewatch:issue-66-prefix-length-sample sneak/routewatch:issue-55-route-per-prefix-and-peer sneak/routewatch:issue-65-batch-write-off-mutex sneak/routewatch:next sneak/routewatch:issue-70-timeout-504 sneak/routewatch:issue-27-cached-stats sneak/routewatch:prod
...
pull from: sneak/routewatch:issue-52-revendor-prompts
Branches Tags
sneak/routewatch:issue-52-revendor-prompts sneak/routewatch:issue-66-prefix-length-sample sneak/routewatch:issue-55-route-per-prefix-and-peer sneak/routewatch:issue-65-batch-write-off-mutex sneak/routewatch:next sneak/routewatch:issue-70-timeout-504 sneak/routewatch:main sneak/routewatch:issue-27-cached-stats sneak/routewatch:prod
21 Commits
issue-27-cached-stats ... issue-52-revendor-prompts
Author SHA1 Message Date
sneak 793c75751d Re-vendor the shared files from sneak/prompts at dd4027b (closes #52)
check / check (push) Waiting to run
Details 
Takes .golangci.yml, .dockerignore, .gitignore, .editorconfig, the
workflow and REPO_POLICIES.md from sneak/prompts dd4027b, keeping this
repository's own entries after the shared content. Lint (golangci-lint
v2.14.0) and test become Dockerfile phases that script/lint and
script/test build alone and uncached; the image build depends on both,
and script/cibuild bootstraps, runs script/check, then builds the image.
The agent guidance file moves to AGENTS.md. The Go code is fixed for
what the new configuration finds; reads that iterate rows now also
report an iteration error.

Model: opus-5-5
 2026-10-14 20:03:00 +00:00
clawbot e0b9c2685f Reconnect when RIS Live ends or stalls the stream (closes #53)
check / check (push) Waiting to run
Details 
When RIS Live ended the response normally, the reconnect loop stopped for good and the health check kept reporting the feed connected; a connection that stayed open but sent nothing blocked forever. Now only `Stop` ends the loop: any other end clears the connected flag and reconnects with the existing backoff. If no line arrives for 60 seconds, including while waiting for RIS Live to answer the request, the streamer drops the connection and reconnects. Tests use a local fake server. README "Reconnection" describes both.

Model: opus-5-5
 2026-10-14 19:42:57 +02:00
clawbot ad4c3d91f1 Show the prefix page's origin names from the prefix query (closes #67)
check / check (push) Waiting to run
Details 
The prefix page looked up each origin AS's details only to show its handle and description, and that lookup reads every live route the AS originates. `GetPrefixDetailsContext` already selected the handle and description and dropped them; it now returns them keyed by AS number, and the page uses them. The origin list now also shows the description after the handle, which it never rendered. The JSON prefix answer is unchanged. The test serves the page from a store whose AS lookup always fails.

Model: opus-5-5
 2026-10-14 19:08:58 +02:00
clawbot b4dee48401 Exit with status 1 when the web server cannot listen (closes #73)
check / check (push) Waiting to run
Details 
`Server.Start` opened its port inside a goroutine and returned nil at once, so a port already in use was only logged and the daemon kept running with no web server. It now opens the listener before returning and returns that error; serving runs in the background on the open listener. `CLIEntry` turns an error from `Run` into an fx shutdown with exit status 1, after the normal stop sequence, so a container restart policy can act. A serving error after a successful listen is still only logged.

Model: opus-5-5
 2026-10-14 18:26:04 +02:00
clawbot 616f13071f Parse the prefix on the prefix pages before looking it up (closes #71)
check / check (push) Waiting to run
Details 
`/prefix/{prefix}/{len}` and `/api/v1/prefix/{prefix}/{len}` passed the URL text to the lookup as typed: a prefix that does not parse answered 500, and a valid prefix written in another form (`2001:DB8:0:0::/32`, `192.0.2.1/24`) answered 404. Both now parse the prefix with `net/netip`, mask it and look up the form the tables store; one that does not parse answers 400. The answer shows the stored form of the prefix. The test drives both pages through the router against a real database.

Model: opus-5-5
 2026-10-14 17:24:02 +02:00
clawbot 521c632087 Ignore routes shorter than /8 or /16 in the IP lookup (closes #57)
check / check (push) Waiting to run
Details 
The IP lookup tried every mask length down to /0, so a default route or a very short route such as `2000::/3` was reported as the owner of any address it covered until a more specific route had been seen. The lookup now stops at /8 for IPv4 and /16 for IPv6 (named constants); an address covered only by shorter routes gets "no route found". The tables still store short routes. The test checks each limit and the length one bit shorter.

Model: opus-5-5
 2026-10-14 17:23:51 +02:00
clawbot 77a708d819 Apply each peer's last update for a prefix in a batch (closes #54)
check / check (push) Waiting to run
Details 
The prefix handler kept one update per prefix in each batch across all peers, chosen by a timestamp cut to whole seconds, so most peers' announcements and withdrawals were lost and the first update in a second won. A batch now keeps the last update per prefix and peer in feed order. A message's withdrawals are queued before its announcements, so a prefix both withdrawn and announced in one message keeps its route (RFC 4271 section 4.3). Tests run each case through the handler into a real database and check the in-memory counts.

Model: opus-5-5
 2026-10-14 17:09:13 +02:00
clawbot a43935308e Count every reconnect in reconnect_count (closes #68)
check / check (push) Waiting to run
Details 
`reconnect_count` in `/api/v1/stats` stayed 0 however often the stream reconnected: `SetConnected` counted a reconnect only while the stream was still marked connected, and the streamer always marks it disconnected first. It now counts one reconnect for every connection after the first, recognised by the zero `connectedSince` before the first connection. A test connects and disconnects twice and expects 2.

Model: opus-5-5
 2026-10-14 15:59:15 +02:00
sneak 00e51d284f Merge pull request 'Next milestone: production memory under 5 GiB, ready for upaas' (#6) from next into main
check / check (push) Waiting to run
Details 
Reviewed-on: #6
 2026-10-06 00:34:48 +02:00
clawbot a9b68608c6 Serve the /api/v1/stats prefix distribution from memory (closes #30)
check / check (push) Successful in 3m19s
Details 
The prefix distribution was the last query on the stats path. It read one index entry per live route on every request, so on a large database it took the whole 4-second deadline and /api/v1/stats answered 500.

The distribution now lives next to the in-memory counts: seeded once at startup from the same query, then kept exact by every live-route write. A new route whose prefix had no live route adds one at its mask length; a delete that leaves a prefix with no live route takes one away. Each check is one lookup on the prefix index, within the same write. The single-route delete now runs its delete and that lookup in one transaction.

Model: opus-5-5
 2026-10-03 18:43:58 +02:00
clawbot 44a5f4cdca Create new databases with auto_vacuum incremental (closes #43)
check / check (push) Successful in 2m50s
Details 
SQLite only accepts auto_vacuum before the database file is first written. The connection switched to WAL first, which writes the file, so the PRAGMA auto_vacuum in Initialize came too late and was ignored. The setting now goes in the connection string, which the driver applies on open before the journal mode, and the late PRAGMA is removed.

Vacuum now reads every row PRAGMA incremental_vacuum returns: SQLite frees one page per row, and the single step ExecContext took freed only one page per call.

Tests check that every pooled connection sees auto_vacuum incremental on a new database and that one Vacuum call frees every page left by deleting routes.

Model: opus-5-5
 2026-10-03 17:09:29 +02:00
clawbot 32704c601e Look up IP addresses by prefix at each mask length (closes #48)
check / check (push) Successful in 3m13s
Details 
Looking up an address on /ip/ and /api/v1/ip/ read every IPv6 live route and about half of the IPv4 range index, so on a day-sized database it passed the 30-second request timeout. The lookup is now one function for both families: from the longest mask length down, it looks up the address's network prefix on the existing prefix index, and the first live route wins.

The feed sends IPv6 withdrawals uncompressed, so they never matched the stored compressed prefix and never removed a route. Prefixes are now stored in the text form net/netip prints, so IPv6 withdrawals take effect.

ip_start, ip_end and GetASInfoForIP are removed; nothing read them. A database from before this change must be deleted.

Model: opus-5-5
 2026-10-03 16:04:38 +02:00
clawbot 7322f936e4 Stamp the git tag or short commit in a plain docker build (closes #46)
check / check (push) Successful in 3m6s
Details 
A plain `docker build .` stamped `unknown` into the page footer:
.dockerignore left out .git, and the Dockerfile built without the -X
flags. The context now carries .git without .git/config, which can hold
a credential. The build stage takes the VERSION build argument when one
is given, otherwise `git describe --tags --always`, fails the build if
.git is present and no version comes out, and stamps it together with
the full commit the footer links to. `make build` now uses
`git describe` as well, and script/docker is the current shared copy.

Model: opus-5-5
 2026-10-02 08:17:35 +02:00
clawbot 6422d9fa0c Container makes its data directory usable itself (closes #42)
check / check (push) Successful in 8s
Details 
sneak's standing rule: the container makes its data directory usable itself, with no step on the host. entrypoint.sh now creates /var/lib/berlin.sneak.app.routewatch if it is missing and stops the start when any step fails (set -euo pipefail); before, a failed cd went on to change the ownership of whatever directory the script was in, and a failed chown still started the daemon. Taking ownership of the directory and switching to the routewatch user through setpriv are unchanged. The README's upaas volume line now says only which path to mount.

The empty-directory and other-uid cases were run by hand on the built image with upaas-style bind mounts, not added as an automated test.

Model: opus-5-5
 2026-09-29 12:23:42 +02:00
clawbot 057e0bd9a9 Add a .dockerignore (closes #39)
check / check (push) Successful in 6s
Details 
Adds a .dockerignore, which REPO_POLICIES.md lists among the files every repo has. Until now every docker build sent the whole working tree as its build context, and the image's source archive is made from that context. It now keeps out .git, local build and test output, archives, local databases, .env and a local Go workspace. Every tracked file stays in, so the format check, the linter, the tests, the build and the source archive read the same files as before.

Without .git in the context the binary's build info records no git revision; nothing in routewatch reads it.

Model: opus-5-5
 2026-09-29 07:42:20 +02:00
clawbot 87a3bc115e README: license and author up front; bring TODO.md up to date (closes #38)
check / check (push) Successful in 3m14s
Details 
Brings README.md and TODO.md in line with REPO_POLICIES.md and with what is on next. The README's first line now names routewatch, what it is, its MIT licence and its author, @sneak; the License section says MIT and links LICENSE; a new Author section names @sneak.

TODO.md's Status, Next Step and Future Steps now describe next as it is: it waits for sneak to merge the milestone PR, after which the upaas deploy and the run under a real 5 GiB limit are his. Completed Steps gains a line for each issue that landed without one.

Docs only. make fmt formats only Go here, so the markdown was wrapped by hand.

Model: opus-5-5
 2026-09-29 07:06:30 +02:00
clawbot d28a59023a Add MIT LICENSE (closes #1)
check / check (push) Successful in 4m5s
Details 
Adds LICENSE with the MIT licence sneak chose for routewatch, copied byte for byte from sneak/webhooker's LICENSE: copyright 2026, Jeffrey Paul. The rest of the issue, .editorconfig and the Gitea check workflow, was already on next, and the README already points at LICENSE.

Judgement call: our repos write the copyright holder in several different ways and sneak named no form; webhooker's gives his full name and email address, the most explicit of them.

Model: opus-5-5
 2026-09-29 05:05:35 +02:00
clawbot 1ac24669d3 Stop the streamer without sending on closed queues (closes #34)
check / check (push) Successful in 2m38s
Details 
Stopping the daemon while the RIS Live feed was flowing could panic with "send on closed channel" and skip the rest of the shutdown. Stop cancels the stream and closes the handler queues under the streamer's write lock, but the read loop checked for a stop only before parsing each line. It now checks again under the read lock it already takes just before handing a message to the queues, so it never sends to a closed queue. Stop also clears its cancel function and returns early when there is none, so a second call no longer closes the queues again. A test forces both cases.

Behaviour change: Stop before Start now does nothing.

Model: opus-5-5
 2026-09-28 21:42:33 +02:00
clawbot 6187ac8503 Let the daemon receive docker stop's signal itself (closes #33)
check / check (push) Successful in 2m42s
Details 
entrypoint.sh now switches to the routewatch user (UID 1000) with setpriv instead of runuser. setpriv replaces itself with the daemon, so the daemon is the container's main process and receives docker stop's signal itself. runuser stayed in between, passed the signal on and killed the daemon 2 seconds later, so every stop ended with exit 143. The daemon now gets the whole wait the caller allows, up to its own 60-second limit, and a clean stop exits 0. Taking ownership of the state directory and the MALLOC_ARENA_MAX check still run as root first.

Not fixed here: a stop while the feed is flowing can still panic (#34).

Model: opus-5-5
 2026-09-28 21:08:33 +02:00
clawbot f2a9e90625 Refuse invalid settings at start, health check follows PORT (closes #31)
check / check (push) Successful in 3m17s
Details 
A set but invalid PORT (anything but plain digits from 1 to 65535) or a relative XDG_DATA_HOME now stops the start before the database opens; before, a bad PORT left the daemon running without HTTP. entrypoint.sh refuses a MALLOC_ARENA_MAX that is not a positive whole number, since glibc ignores a bad one silently. The HEALTHCHECK probes the port PORT names, 8080 when unset. PORT is now read in internal/config, so server.New takes the config.

The README gives the real Linux state directory, lists XDG_DATA_HOME, and adds "Running under upaas": port, volume, environment, the 5g memory limit and the health check.

Unverified: the 5g memory limit could not be exercised on the build host.

Model: opus-5-5
 2026-09-28 20:08:42 +02:00
clawbot df9e23d503 Serve /api/v1/stats counts from realtime in-memory counters (closes #27)
check / check (push) Successful in 2m36s
Details 
Realtime in-memory counters seeded at startup and adjusted on every insert, update and delete; no periodic recompute. Independent review passed: #29 (comment)

model: claude-opus-4-8 (implementation and review); merged by claude-fable-5
 2026-09-22 09:41:16 +02:00
67 changed files with 6462 additions and 3477 deletions
Expand all files Collapse all files
.dockerignore
+97
View File
@@ -0,0 +1,97 @@
# .dockerignore does NOT use .gitignore semantics. Docker matches with
# moby/patternmatcher: filepath.Match plus `**`, so `*` does not cross
# `/` and an unprefixed pattern is anchored at the context root. Every
# depth-independent pattern therefore needs `**/`, or `config/.env` and
# `certs/server.key` still ship while this file reads as solved. Only
# genuinely root-anchored entries go unprefixed. Never transplant these
# into .gitignore, where `**/` is wrong.
#
# Matching is case-sensitive, so secrets use character ranges rather
# than an ALL-CAPS twin, which would still miss `Server.Key`.
#
# Extend with this repo's own host-built artifacts, written anchored:
# `/myapp`, never `**/myapp`, which also matches `cmd/myapp/` and
# deletes the package directory from the context.
# .git is sent without its config. Without a VERSION build argument the
# stage that compiles runs `git describe --tags --always` on .git, which
# does not need .git/config; that file can hold a credential, such as a
# password in a remote URL or the token the CI checkout step stores there.
# Each submodule keeps a config with the same exposure in its git directory
# under .git/modules/, nested again for a submodule's own submodules, or in
# its own .git directory when it keeps one.
# KNOWN GAP: a submodule whose name has a `config` segment (`config`,
# `deploy/config`, `config/lib`) loses its whole git directory, because
# `**/.git/modules/**/config` also matches that segment's directory
# under .git/modules/. Go's version stamping then fails the build;
# nothing leaks. Name such a submodule without that segment:
# `git submodule add --name`.
**/.git/config
**/.git/modules/**/config
# Agent scratch: one full checkout of the repo per in-flight agent.
# Anchored because it occurs once where agents run at the repo root.
# KNOWN GAP: a repo running agents in subdirectories still ships
# `services/api/.claude/` and must add its own anchored entry.
.claude
# Environment files. `*.env` covers bare `.env` and the `prod.env`
# convention. Re-include a committed template with a negation if the
# build needs one: `!docs/example.env`.
**/*.[eE][nN][vV]
**/.[eE][nN][vV].*
**/.[eE][nN][vV][rR][cC]
# Private keys and the bundles carrying them. Public certificates
# (*.crt, *.cer) are deliberately absent: they are legitimate inputs.
**/*.[pP][eE][mM]
**/*.[kK][eE][yY]
**/*.[pP]12
**/*.[pP][fF][xX]
**/[iI][dD]_[rR][sS][aA]
**/[iI][dD]_[dD][sS][aA]
**/[iI][dD]_[eE][cC][dD][sS][aA]
**/[iI][dD]_[eE][cC][dD][sS][aA]_[sS][kK]
**/[iI][dD]_[eE][dD]25519
**/[iI][dD]_[eE][dD]25519_[sS][kK]
# Dependencies: restored inside the image, never copied in.
**/node_modules
# OS metadata.
**/.DS_Store
**/Thumbs.db
# Editor state: never a build input, and it churns COPY.
**/*.swp
**/*.swo
**/*~
**/*.bak
**/.idea
**/.vscode
**/*.sublime-*
# This repository's host-built artifacts: `make build` writes bin/, `go
# build` in the root the binaries, `make run` log.txt, `make asupdate` its
# .tmp file; test binaries, coverage profiles and source archives.
/bin
/routewatch
/asinfo-gen
/streamdumper
/log.txt
/out
/pkg/asinfo/asdata.json
**/*.tar.zst
**/*.test
**/*.out
**/*.tmp
# Local databases. The image carries a source archive of the whole build
# context, so these would otherwise ship inside it.
**/*.db
**/*.db-journal
**/*.db-wal
# A local Go workspace points at directories outside the build context.
/go.work
/go.work.sum
.editorconfig
+3
View File
@@ -10,3 +10,6 @@ insert_final_newline = true
[Makefile] [Makefile]
indent_style = tab indent_style = tab
[*.go]
indent_style = tab
.gitea/workflows/check.yml
+1 -1
View File
@@ -4,6 +4,6 @@ jobs:
check: check:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
# actions/checkout v4.2.2, 2026-02-28 # actions/checkout v4.2.2, 2026-02-22
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- run: script/cibuild - run: script/cibuild
.gitignore
+63 -27
View File
@@ -1,41 +1,77 @@
# Binaries for programs and plugins # OS
.DS_Store
Thumbs.db
# Editors
*.swp
*.swo
*~
*.bak
.idea/
.vscode/
*.sublime-*
# Agent scratch (worktrees of this repo, created and destroyed by
# in-flight tooling). Unanchored: .gitignore patterns already match at
# every depth, so no prefix is wanted here. This is not a .dockerignore
# entry and must not be given a `**/` prefix on the way into one.
.claude/
# Node
node_modules/
# Secrets. Unanchored like every entry above, so each matches at every
# depth. Matching is case-sensitive on Linux, so names use character
# ranges rather than a lowercase form that misses `Server.Key`.
# Environment files. `*.env` covers bare `.env` and the `prod.env`
# convention. Only the templates `example.env` and `sample.env` are
# re-included below. A repository that commits any other template adds
# its own negation after these lines, for example `!.env.example`.
*.[eE][nN][vV]
.[eE][nN][vV].*
.[eE][nN][vV][rR][cC]
!example.env
!sample.env
# Private keys and the bundles carrying them.
*.[pP][eE][mM]
*.[kK][eE][yY]
*.[pP]12
*.[pP][fF][xX]
[iI][dD]_[rR][sS][aA]
[iI][dD]_[dD][sS][aA]
[iI][dD]_[eE][cC][dD][sS][aA]
[iI][dD]_[eE][cC][dD][sS][aA]_[sS][kK]
[iI][dD]_[eE][dD]25519
[iI][dD]_[eE][dD]25519_[sS][kK]
# This repository's own entries, kept after the canonical content above.
# Go: logs, test binaries, coverage profiles, and the binaries `make build`
# and `go build` in the repository root write.
*.log
*.out
*.test
*.exe *.exe
*.exe~ *.exe~
*.dll *.dll
*.so *.so
*.dylib *.dylib
/bin/ /bin/
/routewatch
.DS_Store /asinfo-gen
log.txt /streamdumper
# Test binary, built with `go test -c`
*.test
# Output of the go coverage tool, specifically when used with LiteIDE
*.out
# Dependency directories (remove the comment below to include it)
# vendor/
# Go workspace file
go.work go.work
go.work.sum go.work.sum
# env file # Local databases, temporary files and debug output.
.env
# Database files
*.db *.db
*.db-journal *.db-journal
*.db-wal *.db-wal
# Temporary files
*.tmp *.tmp
# Raw AS data (we only commit the gzipped version)
pkg/asinfo/asdata.json
# Debug output files
out
log.txt log.txt
out
# Raw AS data (only the gzipped version is committed).
pkg/asinfo/asdata.json
.golangci.yml
+88 -81
View File
@@ -1,92 +1,99 @@
version: "2" version: "2"
# Config schema uses the golangci-lint v2 layout (settings live under
# linters.settings, not top-level linters-settings) so that the
# thresholds below are actually applied by golangci-lint >= v2.
run: run:
go: "1.24" timeout: 5m
tests: false modules-download-mode: readonly
linters: linters:
default: all
enable: enable:
# Additional linters requested # Successor to the deprecated gomodguard. Named explicitly, rather than
- testifylint # Checks usage of github.com/stretchr/testify # left to `default: all`, because it carries the module policy below.
- usetesting # usetesting is an analyzer that detects using os.Setenv instead of t.Setenv since Go 1.17 - gomodguard_v2
# - tagliatelle # Disabled: we need snake_case for external API compatibility disable:
- nlreturn # nlreturn checks for a new line before return and branch statements # Genuinely incompatible with project patterns
- nilnil # Checks that there is no simultaneous return of nil error and an invalid value - exhaustruct # Requires all struct fields
- nestif # Reports deeply nested if statements - exhaustruct_v5 # Requires all struct fields (successor to exhaustruct)
- mnd # An analyzer to detect magic numbers - godot # Requires comments to end with periods
- lll # Reports long lines - wrapcheck # Too verbose for internal packages
- intrange # intrange is a linter to find places where for loops could make use of an integer range - varnamelen # Short names like db, id are idiomatic Go
- gochecknoglobals # Check that no global variables exist # Deprecated: the warning is attached to the old name, so it is
# silenced by disabling that name, not by enabling the successor.
# Default/existing linters that are commonly useful - wsl # Deprecated, replaced by wsl_v5
- govet - gomodguard # Deprecated, replaced by gomodguard_v2
- errcheck settings:
- staticcheck lll:
- unused line-length: 88
- ineffassign funlen:
- misspell lines: 80
- revive statements: 50
- gosec cyclop:
- unconvert max-complexity: 15
- unparam dupl:
threshold: 100
linters-settings: depguard:
lll: # Test-support code must not be compiled into the shipped binary. A
line-length: 120 # test-support package exists to hand a test privileges the program
# itself must never have, so a file that is not a test must not import
nestif: # one. Test files, and the files inside a package whose directory name
min-complexity: 4 # ends in `test`, are where that code belongs, and are exempt.
#
nlreturn: # The deny list below is the one part of this file a repository is
block-size: 2 # expected to extend, and the only part it may. depguard matches an
# import path against a list of prefixes, so it cannot be told "any path
revive: # whose last segment ends in test"; a repository's own test-support
rules: # packages have to be named here one at a time, by full import path,
- name: var-naming # under a module path that differs from repository to repository. Add
arguments: # them; change nothing else.
- []
- []
- "upperCaseConst=true"
tagliatelle:
case:
rules: rules:
json: snake test-support:
yaml: snake list-mode: lax
xml: snake files:
bson: snake - "$all"
- "!$test"
testifylint: - "!**/*test/**"
enable-all: true deny:
- pkg: net/http/httptest
usetesting: {} desc: >-
Test-support code belongs in test files and in packages whose
directory name ends in test, not in the shipped binary.
# Only decisions already recorded in the Go package defaults are
# listed here. Every entry matches the module path exactly.
gomodguard_v2:
blocked:
- module: github.com/rs/zerolog
recommendations:
- log/slog
reason: "Structured logging is stdlib log/slog."
# One entry per pre-fork module path, because the later releases
# are separate paths. A prefix match would be shorter but would
# also reach github.com/go-redis/redismock, the test double for
# the successor these entries recommend.
- module: github.com/go-redis/redis
recommendations:
- github.com/redis/go-redis/v9
reason: "Pre-fork module; use the maintained go-redis v9."
- module: github.com/go-redis/redis/v7
recommendations:
- github.com/redis/go-redis/v9
reason: "Pre-fork module; use the maintained go-redis v9."
- module: github.com/go-redis/redis/v8
recommendations:
- github.com/redis/go-redis/v9
reason: "Pre-fork module; use the maintained go-redis v9."
- module: github.com/sergi/go-diff
recommendations:
- github.com/aymanbagabas/go-udiff
reason: "No unified diff output; use go-udiff."
- module: github.com/hexops/gotextdiff
recommendations:
- github.com/aymanbagabas/go-udiff
reason: "Unmaintained fork; use go-udiff."
issues: issues:
max-issues-per-linter: 0 max-issues-per-linter: 0
max-same-issues: 0 max-same-issues: 0
exclude-rules:
# Exclude unused parameter warnings for cobra command signatures
- text: "parameter '(args|cmd)' seems to be unused"
linters:
- revive
# Allow ALL_CAPS constant names
- text: "don't use ALL_CAPS in Go names"
linters:
- revive
# Allow snake_case JSON tags for external API compatibility
- path: "internal/types/ris.go"
linters:
- tagliatelle
# Allow snake_case JSON tags for database models
- path: "internal/database/models.go"
linters:
- tagliatelle
# Allow generic package name for types that define data structures
- path: "internal/types/"
text: "avoid meaningless package names"
linters:
- revive
CLAUDE.md → AGENTS.md
View File
Dockerfile
+50 -24
View File
@@ -1,22 +1,37 @@
# Lint stage — fast feedback on formatting and lint issues. # Lint phase, built alone by script/lint. The linter is invoked directly
# The golangci-lint image bundles Go, gcc and make, so it can run go vet on # rather than through `make lint`, which is itself a docker build and
# the CGO sqlite package and golangci-lint without extra installs. # would recurse into a daemon that does not exist in a build step. The
# golangci/golangci-lint:v2.7.2 (Go 1.25.5), 2026-09-21 # image is Debian-based and has Go and gcc, so it can check the cgo
FROM golangci/golangci-lint@sha256:5d6d5c70a61f1356adfd9dd6316ce286799fefc9d743421356ff1b00842368ba AS lint # SQLite package without extra installs.
# golangci/golangci-lint:v2.14.0 (Go 1.27.0), 2026-10-14
FROM golangci/golangci-lint@sha256:ad862ba6b3798cbe0fd9fd7408d498fd74fbd2623a92406b2fd3898faf0bf98f AS lint
WORKDIR /src WORKDIR /src
COPY go.mod go.sum ./ COPY go.mod go.sum ./
RUN go mod download RUN go mod download
COPY . . COPY . .
RUN golangci-lint run --config .golangci.yml ./...
RUN make fmt-check # Test phase, built alone by script/test. -race needs cgo and so a C
RUN make lint # compiler, which the Debian Go image ships. -short skips the live RIPE RIS
# feed test, so the suite needs no network. The rerun with -v only shows
# details: the build fails however it ends, because the first run failed.
# golang:1.24-bookworm, 2026-09-21
FROM golang@sha256:1a6d4452c65dea36aac2e2d606b01b4a029ec90cc1ae53890540ce6173ea77ac AS test
WORKDIR /src
COPY go.mod go.sum ./
RUN go mod download
COPY . .
RUN go test -short -timeout 90s -race -cover ./... || \
{ echo "--- Rerunning with -v for details ---"; \
go test -short -timeout 90s -race -v ./...; exit 1; }
# Build stage # Build stage. Nothing is wanted from the lint and test phases; the copies
# are what make BuildKit build them first, so this stage cannot run unless
# lint and test passed.
# golang:1.24-bookworm, 2026-09-21 # golang:1.24-bookworm, 2026-09-21
FROM golang@sha256:1a6d4452c65dea36aac2e2d606b01b4a029ec90cc1ae53890540ce6173ea77ac AS builder FROM golang@sha256:1a6d4452c65dea36aac2e2d606b01b4a029ec90cc1ae53890540ce6173ea77ac AS builder
COPY --from=lint /src/go.sum /dev/null
COPY --from=test /src/go.sum /dev/null
# Install build dependencies (zstd for archive, gcc for CGO/sqlite3) # Install build dependencies (zstd for archive, gcc for CGO/sqlite3)
RUN apt-get update && apt-get install -y --no-install-recommends \ RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -25,10 +40,10 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
libc6-dev \ libc6-dev \
&& rm -rf /var/lib/apt/lists/* && rm -rf /var/lib/apt/lists/*
WORKDIR /src # A tar-stream context keeps the sender's file owners, which git refuses.
RUN git config --system --add safe.directory /src
# Force BuildKit to run the lint stage before compiling or testing. WORKDIR /src
COPY --from=lint /src/go.sum /dev/null
# Copy everything # Copy everything
COPY . . COPY . .
@@ -36,12 +51,23 @@ COPY . .
# Vendor dependencies (must be after copying source) # Vendor dependencies (must be after copying source)
RUN go mod download && go mod vendor RUN go mod download && go mod vendor
# Run the test suite in the build stage: -race needs cgo and the C compiler # Build the binary with CGO enabled (required for sqlite3). The version the
# installed above. The suite is offline (the live-feed test is opt-in). # page footer shows is the VERSION build argument when one is given, otherwise
RUN make test # `git describe --tags --always` of the .git in the build context (git comes
# with this image): the tag on a tagged commit, tag-N-gHASH after one, the
# Build the binary with CGO enabled (required for sqlite3) # short commit when no tag is reachable. A context that carries .git and still
RUN CGO_ENABLED=1 GOOS=linux go build -o /routewatch ./cmd/routewatch # yields no version fails the build. The footer links to the full commit.
ARG VERSION
RUN version="${VERSION:-$(git describe --tags --always || echo unknown)}"; \
if [ -e .git ] && { [ -z "$version" ] || [ "$version" = dev ] || \
[ "$version" = unknown ]; }; then \
echo "no version could be derived although the build context carries .git" >&2; \
exit 1; \
fi; \
CGO_ENABLED=1 GOOS=linux go build -o /routewatch -ldflags "\
-X git.eeqj.de/sneak/routewatch/internal/version.GitRevision=$(git rev-parse --verify HEAD || echo unknown) \
-X git.eeqj.de/sneak/routewatch/internal/version.GitRevisionShort=$version" \
./cmd/routewatch
# Create source archive with vendored dependencies # Create source archive with vendored dependencies
RUN tar --zstd -cf /routewatch-source.tar.zst \ RUN tar --zstd -cf /routewatch-source.tar.zst \
@@ -79,8 +105,8 @@ RUN chown -R routewatch:routewatch /app
ENV XDG_DATA_HOME=/var/lib ENV XDG_DATA_HOME=/var/lib
# Cap the Go heap at 1.5 GiB so the runtime collects harder before the # Cap the Go heap at 1.5 GiB so the runtime collects harder before the
# container's memory limit is reached. runuser preserves this the way it does # container's memory limit is reached. setpriv in the entrypoint preserves this
# XDG_DATA_HOME above. # the way it does XDG_DATA_HOME above.
ENV GOMEMLIMIT=1536MiB ENV GOMEMLIMIT=1536MiB
# Cap glibc's malloc arenas. The SQLite C library allocates and frees millions # Cap glibc's malloc arenas. The SQLite C library allocates and frees millions
@@ -96,8 +122,8 @@ EXPOSE 8080
COPY ./entrypoint.sh /entrypoint.sh COPY ./entrypoint.sh /entrypoint.sh
# Health check using the health endpoint # Health check using the health endpoint, on the port PORT names
HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \ HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
CMD curl -sf http://localhost:8080/.well-known/healthcheck.json || exit 1 CMD curl -sf "http://localhost:${PORT:-8080}/.well-known/healthcheck.json" || exit 1
ENTRYPOINT ["/bin/bash", "/entrypoint.sh" ] ENTRYPOINT ["/bin/bash", "/entrypoint.sh" ]
LICENSE
+21
View File
@@ -0,0 +1,21 @@
MIT License
Copyright (c) 2026 Jeffrey Paul <sneak@sneak.berlin>
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Makefile
+1 -1
View File
@@ -2,7 +2,7 @@ export DEBUG = routewatch
# Git revision for version embedding # Git revision for version embedding
GIT_REVISION := $(shell git rev-parse HEAD 2>/dev/null || echo "unknown") GIT_REVISION := $(shell git rev-parse HEAD 2>/dev/null || echo "unknown")
GIT_REVISION_SHORT := $(shell git rev-parse --short HEAD 2>/dev/null || echo "unknown") GIT_REVISION_SHORT := $(shell git describe --tags --always 2>/dev/null || echo "unknown")
VERSION_PKG := git.eeqj.de/sneak/routewatch/internal/version VERSION_PKG := git.eeqj.de/sneak/routewatch/internal/version
LDFLAGS := -X $(VERSION_PKG).GitRevision=$(GIT_REVISION) -X $(VERSION_PKG).GitRevisionShort=$(GIT_REVISION_SHORT) LDFLAGS := -X $(VERSION_PKG).GitRevision=$(GIT_REVISION) -X $(VERSION_PKG).GitRevisionShort=$(GIT_REVISION_SHORT)
README.md
+59 -16
View File
@@ -1,6 +1,9 @@
# RouteWatch # RouteWatch
RouteWatch is a real-time BGP routing table monitor that streams BGP UPDATE messages from the RIPE RIS Live service, maintains a live routing table in SQLite, and provides HTTP APIs for querying routing information. RouteWatch is an MIT-licensed Go daemon by @sneak that monitors the BGP routing
table in real time: it streams BGP UPDATE messages from the RIPE RIS Live
service, maintains a live routing table in SQLite, and provides HTTP APIs for
querying routing information.
## Features ## Features
@@ -138,8 +141,15 @@ routewatch/
- **Batched Writes**: All database operations are batched for performance - **Batched Writes**: All database operations are batched for performance
- **Backpressure**: Probabilistic message dropping when queues exceed 50% capacity - **Backpressure**: Probabilistic message dropping when queues exceed 50% capacity
- **Graceful Shutdown**: 60-second timeout, flushes all pending batches - **Graceful Shutdown**: 60-second timeout, flushes all pending batches
- **Reconnection**: Exponential backoff (5s-320s) with reset after 30s of stable connection - **Reconnection**: Any end of the stream other than shutdown, including RIS Live
- **IPv4 Optimization**: IP ranges stored as uint32 for O(1) lookups ending the response normally, marks the feed disconnected and reconnects with
exponential backoff (5s-320s). The backoff resets after a stream that lasted
more than 30s. A stream with no line for 60s is dropped and reconnected.
- **IP Lookup**: the most specific live route is found by looking up the
address's prefix at each mask length, longest first, in the prefix index
(at most 25 lookups for IPv4, 113 for IPv6); prefixes are stored in the
text form Go's `net/netip` prints. Routes shorter than /8 for IPv4 or /16
for IPv6, such as a default route, are not used.
### Database Schema ### Database Schema
@@ -151,7 +161,7 @@ prefixes_v6(id, prefix, mask_length, first_seen, last_seen)
-- Live routing tables (one per IP version) -- Live routing tables (one per IP version)
live_routes_v4(id, prefix, mask_length, origin_asn, peer_ip, as_path, live_routes_v4(id, prefix, mask_length, origin_asn, peer_ip, as_path,
next_hop, last_updated, v4_ip_start, v4_ip_end) next_hop, last_updated)
live_routes_v6(id, prefix, mask_length, origin_asn, peer_ip, as_path, live_routes_v6(id, prefix, mask_length, origin_asn, peer_ip, as_path,
next_hop, last_updated) next_hop, last_updated)
@@ -166,14 +176,21 @@ Configuration is handled via environment variables and OS-specific paths:
| Variable | Default | Description | | Variable | Default | Description |
|----------|----------|-------------| |----------|----------|-------------|
| `PORT` | `8080` | HTTP server port | | `PORT` | `8080` | HTTP server port, a whole number from 1 to 65535 |
| `DEBUG` | (empty) | Set to `routewatch` for debug logging | | `DEBUG` | (empty) | Set to `routewatch` for debug logging |
| `XDG_DATA_HOME` | `/var/lib` (in the Docker image) | Base of the state directory; must be an absolute path |
| `GOMEMLIMIT` | `1536MiB` (in the Docker image) | Go soft memory limit; see Memory | | `GOMEMLIMIT` | `1536MiB` (in the Docker image) | Go soft memory limit; see Memory |
| `MALLOC_ARENA_MAX` | `2` (in the Docker image) | glibc malloc arena cap; see Memory | | `MALLOC_ARENA_MAX` | `2` (in the Docker image) | glibc malloc arena cap, a positive whole number; see Memory |
A variable that is set to an invalid value stops the start with an error and a
non-zero exit. An empty variable counts as unset.
State directory (database location): State directory (database location):
- macOS: `~/Library/Application Support/routewatch/` - macOS: `~/Library/Application Support/routewatch/`
- Linux: `/var/lib/routewatch/` or `~/.local/share/routewatch/` - Linux: `/var/lib/berlin.sneak.app.routewatch/` when running as root,
otherwise `$XDG_DATA_HOME/berlin.sneak.app.routewatch/` (with `XDG_DATA_HOME`
unset, `~/.local/share/berlin.sneak.app.routewatch/`). In the Docker image
this is `/var/lib/berlin.sneak.app.routewatch/`.
## Memory ## Memory
@@ -220,6 +237,22 @@ What happens at each limit:
With `DEBUG=routewatch` the daemon logs a `System stats` line every 60 seconds With `DEBUG=routewatch` the daemon logs a `System stats` line every 60 seconds
with the goroutine count and Go memory figures. with the goroutine count and Go memory figures.
## Running under upaas
What the [upaas](https://git.eeqj.de/sneak/upaas) app needs:
- Container port: `8080`.
- Volume: one, at container path `/var/lib/berlin.sneak.app.routewatch`.
- Environment: nothing is required. Leave `XDG_DATA_HOME`, `GOMEMLIMIT` and
`MALLOC_ARENA_MAX` at the image's values. `DEBUG=routewatch` is optional and
adds the `System stats` memory line to the log.
- Memory Limit: `5g`, the 5 GiB limit from Memory above. upaas sets no swap
limit, so on a host with swap Docker allows the same amount of swap again.
- Health check: the image's `HEALTHCHECK` requests
`/.well-known/healthcheck.json` on the container port. upaas reads the
container's health 60 seconds after a deploy and fails the deploy unless it
is `healthy`.
## Development ## Development
```bash ```bash
@@ -244,29 +277,35 @@ standard: normalized scripts in `script/` are the entrypoints for the
development workflow, and the Makefile targets are thin shims that call development workflow, and the Makefile targets are thin shims that call
them. We provide: them. We provide:
- `script/bootstrap` — install all development dependencies (go, - `script/bootstrap` — install all development dependencies (git, make, go, Go
golangci-lint, Go module download) module download); warns when docker is missing
- `script/setup` — make a fresh clone ready for development: runs - `script/setup` — make a fresh clone ready for development: runs
`script/bootstrap`, then `script/install-precommit` `script/bootstrap`, then `script/install-precommit`
- `script/projectname` — print the project name (used for the Docker - `script/projectname` — print the project name (used for the Docker
image tag) image tag)
- `script/test` — run the test suite - `script/test` — build the `test` phase alone, which runs
(`go test -short -timeout 30s -race -cover ./...`, verbose rerun on `go test -short -timeout 90s -race -cover ./...` (verbose rerun on failure).
failure). The `-short` flag skips the live-network integration test so the The `-short` flag skips the live-network integration test so the run is
default run is deterministic and offline. deterministic and offline.
- `script/lint` — run `go vet ./...` and `golangci-lint run` - `script/lint` — build the `lint` phase alone, which runs golangci-lint
v2.14.0 with `.golangci.yml`
- `script/fmt` — format all code (writes) - `script/fmt` — format all code (writes)
- `script/fmt-check` — check formatting (read-only) - `script/fmt-check` — check formatting (read-only)
- `script/check` — run `script/test`, `script/lint`, and - `script/check` — run `script/test`, `script/lint`, and
`script/fmt-check` `script/fmt-check`
- `script/docker` — build the Docker image tagged via - `script/docker` — build the Docker image tagged via
`script/projectname` `script/projectname`
- `script/cibuild` — CI entrypoint: `docker build .` - `script/cibuild` — CI entrypoint: runs `script/bootstrap` and
`script/check`, then builds the image
- `script/precommit` — pre-commit gate: `go mod tidy` + `go fmt` (must - `script/precommit` — pre-commit gate: `go mod tidy` + `go fmt` (must
not change files), then `script/check` not change files), then `script/check`
- `script/install-precommit` — install the git pre-commit hook that - `script/install-precommit` — install the git pre-commit hook that
runs `script/precommit` runs `script/precommit`
Tests and the linter run only in Docker, as the `test` and `lint` phases of the
`Dockerfile`; the image cannot be built unless both pass. Every `docker build`
in `script/` passes `--no-cache`, so each run executes every check.
The live-network integration test `TestRouteWatchLiveFeed` streams the RIPE The live-network integration test `TestRouteWatchLiveFeed` streams the RIPE
RIS feed for a few seconds and is skipped in short mode. To run it on demand, RIS feed for a few seconds and is skipped in short mode. To run it on demand,
invoke `go test` directly without `-short`: invoke `go test` directly without `-short`:
@@ -274,4 +313,8 @@ invoke `go test` directly without `-short`:
## License ## License
See LICENSE file. MIT. See [`LICENSE`](LICENSE).
## Author
[@sneak](https://sneak.berlin)
REPO_POLICIES.md
+357 -86
View File
@@ -1,6 +1,6 @@
--- ---
title: Repository Policies title: Repository Policies
last_modified: 2026-07-06 last_modified: 2026-10-04
--- ---
This document covers repository structure, tooling, and workflow standards. Code This document covers repository structure, tooling, and workflow standards. Code
@@ -60,17 +60,28 @@ style conventions are in separate documents:
prerequisite since nvm requires bash. yarn is then pinned via prerequisite since nvm requires bash. yarn is then pinned via
`corepack prepare yarn@<version> --activate`. Never install "latest" or "lts"; `corepack prepare yarn@<version> --activate`. Never install "latest" or "lts";
always exact versions. `script/cibuild` runs the CI build: it changes to the always exact versions. `script/cibuild` runs the CI build: it changes to the
repo root and runs `docker build .`; the Gitea workflow calls it. Four further repo root, runs `script/bootstrap`, runs `script/check`, and builds the image
scripts are our own extensions to the standard: `script/check` runs with the version; the Gitea workflow calls it. **`script/cibuild` runs
`script/test`, `script/lint`, and `script/fmt-check`; `script/precommit` is `script/bootstrap` first**, because the workflow checks out the repo and runs
what the git pre-commit hook runs, and it calls `script/check`; nothing else, while `script/fmt-check` runs the formatter on the host: on a
`script/install-precommit` installs the git pre-commit hook (the `make hooks` pristine checkout with nothing installed the run dies there, after the
target shims to it); and `script/projectname` (literally that filename) simply containerised gates have passed. **The bootstrap alone is not enough**:
outputs the project's name. Scripts that need the name call `script/bootstrap` installs node and yarn under nvm and leaves neither on the
`script/projectname` — e.g. `script/docker` assembles its image tag from it — `PATH` of the shell that called it, so a bare `yarn` still exits 127. The host
so those scripts stay byte-identical across all repos. Repo-type-specific entrypoints that need yarn — `script/fmt` and `script/fmt-check` — therefore
pre-commit extras (e.g. `go mod tidy` verification in Go repos) belong in source nvm for the pinned node version before invoking it, exactly as
`script/precommit`, not in the hook itself. Model scripts are at `script/bootstrap`'s own install step does. A runner carrying nothing but
docker and git then gets through `script/check`. Four further scripts are our
own extensions to the standard: `script/check` runs `script/test`,
`script/lint` and `script/fmt-check`; `script/precommit` is what the git
pre-commit hook runs, and it calls `script/check`; `script/install-precommit`
installs the git pre-commit hook (the `make hooks` target shims to it); and
`script/projectname` (literally that filename) simply outputs the project's
name. Scripts that need the name call `script/projectname` — e.g.
`script/docker` assembles its image tag from it — so those scripts stay
byte-identical across all repos. Repo-type-specific pre-commit extras (e.g.
`go mod tidy` verification in Go repos) belong in `script/precommit`, not in
the hook itself. Model scripts are at
`https://git.eeqj.de/sneak/prompts/raw/branch/main/script/<name>`. The README `https://git.eeqj.de/sneak/prompts/raw/branch/main/script/<name>`. The README
must document the provided scripts in an **Entrypoints** section (see the must document the provided scripts in an **Entrypoints** section (see the
README requirements below). README requirements below).
@@ -89,87 +100,198 @@ style conventions are in separate documents:
contributor should be able to understand the entire development workflow by contributor should be able to understand the entire development workflow by
reading the Makefile. reading the Makefile.
- Every repo should have a `Dockerfile`. All Dockerfiles must run `make check` - Every repo should have a `Dockerfile`, and it carries the repo's gates: a
as a build step so the build fails if the branch is not green. For non-server `lint` phase and a `test` phase, with the final stage depending on both so the
repos, the Dockerfile should bring up a development environment and run image cannot be built unless they pass. For non-server repos the final stage
`make check`. For server repos, `make check` should run as an early build brings up a development environment; for server repos it is the runtime image.
stage before the final image is assembled. Dockerfiles install development The gate phases and the build stage start from their pinned base images and
prerequisites by running `script/bootstrap` rather than duplicating installs install what those images lack either inline, as the canonical Go `Dockerfile`
inline; COPY `script/` and the dependency manifests (`package.json` + below does for `git`, or by running `script/bootstrap`, as the `prompts`
`yarn.lock`, `go.mod` + `go.sum`, etc.) before running it so the bootstrap repo's own `Dockerfile` does for its yarn packages. The development
layer stays cached until dependencies change. environment stage installs development prerequisites by running
`script/bootstrap` rather than duplicating its installs inline. A stage that
runs `script/bootstrap` COPYs `script/` and the dependency manifests
(`package.json` + `yarn.lock`, `go.mod` + `go.sum`, etc.) before running it.
- **Dockerfiles must use a separate lint stage for fail-fast feedback.** Go - **Linting and testing run in Docker, as phases of the `Dockerfile`.** There is
repos use a multistage build where linting runs in an independent stage based no separate lint file. `script/lint` and `script/test` each build one phase
on the `golangci/golangci-lint` image (pinned by hash). This stage runs and nothing else:
`make fmt-check` and `make lint` before the full build begins. The build stage
then declares an explicit dependency on the lint stage via
`COPY --from=lint /src/go.sum /dev/null`, which forces BuildKit to complete
linting before proceeding to compilation and tests. This ensures lint failures
surface in seconds rather than minutes, without blocking on dependency
download or compilation in the build stage.
The standard pattern for a Go repo Dockerfile is: ```sh
docker build --no-cache --target lint -t "$(script/projectname)-lint" .
docker build --no-cache --target test -t "$(script/projectname)-test" .
```
**A stage that is not the last one in the file is built only when the final
stage's chain depends on it, or when `--target` names it.** That is why the
two gates are always invoked by name here, and why the final stage carries a
`COPY --from=` of a harmless file from each of them: without that edge a
plain `docker build .` builds the last stage alone and exits 0 having linted
and tested nothing.
**Every `docker build` in `script/` is tagged**, here and in
`script/cibuild` and `script/docker`. An untagged build leaves a dangling
image behind on every invocation, on every developer host and every CI
runner; a tagged one replaces the previous image.
Inside a phase the tool is invoked directly — `golangci-lint`, `go test`,
`eslint`, `prettier` — never through `make lint` or `script/test`, which are
themselves a `docker build` and would recurse into a daemon that does not
exist in a build step. Formatting is the exception and stays on the host:
`script/fmt` writes the working tree, and `script/fmt-check` is its
read-only twin.
**No lint verdict may come from a host invocation of the linter.** On a
shared host golangci-lint reads a result cache keyed on file content rather
than location, so a second checkout of the same content is served the first
one's findings, and a host-global lock in `$TMPDIR` makes concurrent runs
exit non-zero with `parallel golangci-lint is running` — a status a caller
cannot tell from real findings. Both have produced wrong verdicts in this
org, in both directions. A container has its own cache, its own `TMPDIR` and
a digest-pinned binary, so neither is reachable.
- **Any build that runs checks is built with `--no-cache`.** Docker invalidates
a `COPY` layer only when the copied content changes, so on an unchanged tree
the check `RUN` is served from cache, nothing executes, and the build still
exits 0. Every `docker build` in `script/` therefore passes `--no-cache`:
`script/lint`, `script/test`, `script/cibuild` and `script/docker` are the
four, and there is no fifth — `script/check` runs the two gate phases and
`script/fmt-check`, and builds no image of its own. A bare `docker build .` is
not evidence that anything ran: a sub-second build reporting success is a
cache hit, not a result. Never invalidate by pruning — `docker builder prune`
and friends destroy a build cache shared with every other build on the host.
When a check is added or changed, prove it works by planting a defect it must
catch and watching the run fail on it, then revert the defect. A green run
alone shows neither that the check ran nor that it covers what it should.
- **The gate phases are separate stages, and the build stage depends on both.**
The lint phase is based on the `golangci/golangci-lint` image (pinned by
hash), so lint failures surface in seconds rather than after a full compile,
and the test phase is based on the Debian Go image. The canonical Go repo
`Dockerfile`:
```dockerfile ```dockerfile
# Lint stage — fast feedback on formatting and lint issues # Lint phase
# golangci/golangci-lint:v2.x.x, YYYY-MM-DD # golangci/golangci-lint:v2.x.x, YYYY-MM-DD
FROM golangci/golangci-lint@sha256:... AS lint FROM golangci/golangci-lint@sha256:... AS lint
WORKDIR /src WORKDIR /src
COPY go.mod go.sum ./ COPY go.mod go.sum ./
RUN go mod download RUN go mod download
COPY . . COPY . .
RUN make fmt-check RUN golangci-lint run --config .golangci.yml ./...
RUN make lint
# Build stage # Test phase. -race needs cgo and so a C compiler, which the Debian Go
# golang:1.x-alpine, YYYY-MM-DD # image ships and the alpine one does not.
FROM golang@sha256:... AS builder # golang:1.x, YYYY-MM-DD
FROM golang@sha256:... AS test
WORKDIR /src WORKDIR /src
# Force BuildKit to run the lint stage before proceeding
COPY --from=lint /src/go.sum /dev/null
COPY go.mod go.sum ./ COPY go.mod go.sum ./
RUN go mod download RUN go mod download
COPY . . COPY . .
RUN make test RUN go test -timeout 90s -race -cover ./... || \
{ echo "--- Rerunning with -v for details ---"; \
go test -timeout 90s -race -v ./...; exit 1; }
ARG VERSION=dev # Build stage. Nothing is wanted from either phase above; the copies
RUN CGO_ENABLED=0 go build -trimpath \ # are what make BuildKit build them first, so this stage cannot run
-ldflags="-s -w -X main.Version=${VERSION}" \ # unless lint and test passed.
-o /app ./cmd/app/ # golang:1.x-alpine, YYYY-MM-DD
FROM golang@sha256:... AS builder
COPY --from=lint /src/go.sum /dev/null
COPY --from=test /src/go.sum /dev/null
RUN apk add --no-cache git
# A tar-stream context keeps the sender's file owners, which git refuses.
RUN git config --system --add safe.directory /src
WORKDIR /src
COPY go.mod go.sum ./
RUN go mod download
COPY . .
# Runtime stage # The VERSION build arg when one is given, otherwise
# `git describe --tags --always` on the .git in the build context. With
# .git present, a version that is still empty, dev or unknown fails the
# build: git is missing or could not read the checkout.
ARG VERSION
RUN VERSION="${VERSION:-$(git describe --tags --always)}"; \
if [ -e .git ]; then \
case "$VERSION" in ""|dev|unknown) \
echo "version is '$VERSION' although .git is present" >&2; \
exit 1 ;; \
esac; \
fi; \
CGO_ENABLED=0 go build -trimpath \
-ldflags="-s -w -X main.Version=${VERSION}" \
-o /app ./cmd/app/
# Runtime stage, and the last one
FROM alpine@sha256:... FROM alpine@sha256:...
COPY --from=builder /app /usr/local/bin/app COPY --from=builder /app /usr/local/bin/app
ENTRYPOINT ["app"] ENTRYPOINT ["app"]
``` ```
Key points: Key points:
- The lint stage uses the `golangci/golangci-lint` image directly (it - The lint phase uses the `golangci/golangci-lint` image directly (it has
includes both Go and the linter), so there is no need to install the both Go and the linter), so nothing needs installing.
linter separately. - `COPY --from=<phase> /src/go.sum /dev/null` is a no-op copy whose only
- `COPY --from=lint /src/go.sum /dev/null` is a no-op file copy that creates purpose is the ordering edge. BuildKit runs stages in parallel by default,
a stage dependency. BuildKit runs stages in parallel by default; without and a stage nothing depends on is not built at all, so without these two
this line, the build stage would not wait for lint to finish and a lint lines a red gate would not fail the build.
failure might not fail the overall build. - Keep the runtime stage last, and if you add a stage after it, give it the
same two copies. A plain `docker build .` builds the last stage's chain
and nothing else.
- If the project uses `//go:embed` directives that reference build artifacts - If the project uses `//go:embed` directives that reference build artifacts
(e.g. a web frontend compiled in a separate stage), the lint stage must (e.g. a web frontend compiled in a separate stage), the lint phase must
create placeholder files so the embed directives resolve. Example: create placeholder files so the embed directives resolve. Example:
`RUN mkdir -p web/dist && touch web/dist/index.html web/dist/style.css`. `RUN mkdir -p web/dist && touch web/dist/index.html web/dist/style.css`.
The lint stage should not depend on the actual build output — it exists to - If the project requires CGO or system libraries for linting, install them
fail fast. in the lint phase. The `golangci/golangci-lint` image is Debian-based and
- If the project requires CGO or system libraries for linting (e.g. has no `apk`, so install with `apt-get` under the Debian package name
`vips-dev`), install them in the lint stage with `apk add`. (`libvips-dev`, where alpine says `vips-dev`), and delete the package
- The build stage runs `make test` after compilation setup. Tests run in the lists in the same `RUN`, so the layer does not keep them:
build stage, not the lint stage, because they may require compiled
artifacts or heavier dependencies. ```dockerfile
RUN apt-get update \
&& apt-get install -y --no-install-recommends libvips-dev \
&& rm -rf /var/lib/apt/lists/*
```
- `.dockerignore` lets `.git` into the build context. It keeps out every git
`config` at any depth (`**/.git/config`, `**/.git/modules/**/config`): the
repository's own, each submodule's under `.git/modules/`, and that of a
submodule keeping its own `.git` directory. `git describe` does not need
them, and each can hold a credential: a password in a remote URL, or the
token the CI checkout step stores there. A submodule whose name has a
`config` segment (`config`, `deploy/config`, `config/lib`) loses its whole
git directory to `**/.git/modules/**/config`, and Go's version stamping
then fails the build: give it a name without that segment
(`git submodule add --name`). The stage that compiles has `git` (the
Debian Go image has it; an alpine one needs `apk add --no-cache git`) and
takes the version from the `VERSION` build argument when one is given,
otherwise from `git describe --tags --always`. That gives the tag on a
tagged commit; on a later commit, the tag, the number of commits since it
and the short commit (`v1.2.3-4-gabc1234`); and the short commit when no
tag is reachable. The stage that compiles also marks its working directory
safe for git (`git config --system --add safe.directory /src`): a context
sent as a tar stream keeps the sender's file owners, and git refuses a
checkout owned by another user, so the version would come out empty.
`ARG VERSION` has no default, and the build fails if the context carries
`.git` and the version still comes out empty, `dev` or `unknown`. A plain
`docker build .` with no build arguments must succeed; a Dockerfile that
refuses an empty build argument drops that refusal and keeps the argument.
- Every repo should have a Gitea Actions workflow (`.gitea/workflows/`) that - Every repo should have a Gitea Actions workflow (`.gitea/workflows/`) that
runs `script/cibuild` (which runs `docker build .`) on push. Since the runs `script/cibuild` on push, and checks out the repo as its only other step.
Dockerfile already runs `make check`, a successful build implies all checks That script bootstraps, runs the gate phases, and then builds the image, so a
pass. successful run means every check passed; a bare `docker build .` does not
carry the same guarantee, because its gate phases may come from the cache. The
image build is uncached and so runs the gate phases a second time. That is the
price of the rule above, and it is worth paying: the image that ships is built
from a run of its own gates rather than from a cache entry. A separate
workflow limited to `main` by a `branches` list under `on: push` cannot be
checked by review: to try a change to it, add the feature branch to that list
and push, then remove the branch from the list again before merging. Keep any
job in it that publishes behind `if: github.ref_name == 'main'`, so the run
from the feature branch publishes nothing.
- Use platform-standard formatters: `black` for Python, `prettier` for - Use platform-standard formatters: `black` for Python, `prettier` for
JS/CSS/Markdown/HTML, `go fmt` for Go. Always use default configuration with JS/CSS/Markdown/HTML, `go fmt` for Go. Always use default configuration with
@@ -189,14 +311,21 @@ style conventions are in separate documents:
module under test to verify it compiles/parses. There is no excuse for module under test to verify it compiles/parses. There is no excuse for
`make test` to be a no-op. `make test` to be a no-op.
- `make test` must complete in under 20 seconds. Add a 30-second timeout in the - `make test` must complete in under 60 seconds. That is the hard cap, and a
Makefile. suite that exceeds it fails. Under 20 seconds is the target. A suite between
20 and 60 seconds is still green, but the overage must be filed as an
improvement bug against that repo. Add a 90-second timeout to the test
invocation (`go test -timeout 90s`). The backstop deliberately sits above the
hard cap so that it catches a genuinely hung test rather than a merely slow
one.
- **`make test` should use the conditional verbose rerun pattern.** Run tests - **The test command should use the conditional verbose rerun pattern.** Run
without `-v` (verbose) first. If tests fail, automatically rerun with `-v` to tests without `-v` (verbose) first. If tests fail, automatically rerun with
show full output. This keeps CI logs and `docker build` output clean on `-v` to show full output. This keeps CI logs and `docker build` output clean
success (just package/suite summaries) while providing full diagnostic detail on success (just package/suite summaries) while providing full diagnostic
on failure (every test case, every assertion). The general shell pattern: detail on failure (every test case, every assertion). The command lives in the
`test` phase of the `Dockerfile`, since `script/test` builds that phase; the
Makefile form below is the same pattern for any repo-local invocation:
```makefile ```makefile
test: test:
@@ -209,11 +338,26 @@ style conventions are in separate documents:
```makefile ```makefile
test: test:
@go test -timeout 30s -race -cover ./... || \ @go test -count=1 -timeout 90s -race -cover ./... || \
{ echo "--- Rerunning with -v for details ---"; \ { echo "--- Rerunning with -v for details ---"; \
go test -timeout 30s -race -v ./...; exit 1; } go test -count=1 -timeout 90s -race -v ./...; exit 1; }
``` ```
`-count=1` is required on both invocations: it defeats Go's test _result_
cache, so neither run can report a stored pass in place of running the
tests. It leaves the build cache alone, so it costs the runtime of the suite
and no recompilation.
That cache is Go's own, separate from Docker's layer cache. Go stores a
passing result in its cache directory (`GOCACHE`), and when the same tests
run again on unchanged code it prints that result, marked `(cached)`,
without running them. That matters on a developer's machine, where this
target runs and the directory lasts from one run to the next. The `test`
phase of the `Dockerfile` needs no `-count=1`: its base image holds no
result for this repo's tests and nothing before its `go test` step runs a
test, so there is nothing to replay. `--no-cache` (above) is what makes that
step run on an unchanged tree.
Python example: Python example:
```makefile ```makefile
@@ -239,10 +383,84 @@ style conventions are in separate documents:
must be in `.gitignore`. No exceptions. must be in `.gitignore`. No exceptions.
- `.gitignore` should be comprehensive from the start: OS files (`.DS_Store`), - `.gitignore` should be comprehensive from the start: OS files (`.DS_Store`),
editor files (`.swp`, `*~`), language build artifacts, and `node_modules/`. editor files (`.swp`, `*~`), in-repo agent scratch directories (`.claude/`),
Fetch the standard `.gitignore` from language build artifacts, and `node_modules/`. Fetch the standard `.gitignore`
`https://git.eeqj.de/sneak/prompts/raw/branch/main/.gitignore` when setting up from `https://git.eeqj.de/sneak/prompts/raw/branch/main/.gitignore` when
a new repo. setting up a new repo. These patterns are written to `.gitignore`'s own
semantics, in which an unanchored pattern already matches at every depth; they
are not a `.dockerignore` and must not be transplanted into one unmodified.
- **`.dockerignore` does not use `.gitignore` semantics, and copying patterns
across unmodified leaves secrets in the build context.** Docker matches with
`moby/patternmatcher`: `filepath.Match` semantics plus a `**` extension, so
`*` does not cross `/` and a pattern without a leading `**/` is anchored at
the build-context root. A `.dockerignore` listing `.env`, `*.pem` and `*.key`
therefore excludes only the copies at the repository root, while `config/.env`
and `certs/server.key` still reach the context and can land in an image layer
— which is more dangerous than a short file with no secret patterns at all,
because it reads as solved and stops anyone looking. Give every
depth-independent pattern the `**/` prefix and leave only genuinely
root-anchored entries unprefixed: `.claude`, and the repo's own host-built
binary, written `/myapp` and never `**/myapp`, which would also match
`cmd/myapp/` and delete the package directory from the context. Matching is
case-sensitive, and an ALL-CAPS twin per pattern still misses `Server.Key`, so
secret names use character ranges — `**/*.[kK][eE][yY]`, `**/*.[pP][eE][mM]`,
and likewise for `.envrc` and the extensionless SSH keys. Where such a pattern
also catches something the build needs, re-include it with a negation
(`!docs/example.env`); deleting the pattern reopens the exposure for every
other file it covers. Fetch the standard `.dockerignore` from
`https://git.eeqj.de/sneak/prompts/raw/branch/main/.dockerignore` and extend
it with the repo's own artifacts.
- **In-repo agent scratch belongs in both files, written to each file's own
semantics.** `.claude/` holds one worktree per in-flight agent — an entire
additional checkout of the repo — so under `COPY . .` the build context
inflates by a multiple of the repo and another session's unreviewed work can
be copied into an image layer. In `.gitignore` the entry is `.claude/`,
unanchored. In `.dockerignore` it is `.claude`, anchored and with **no** `**/`
prefix, because the prefixed form would also delete any nested directory of
that name from the build. Anchoring carries a known gap that the canonical
`.dockerignore` states in its own comment, since consuming repos receive the
file and not the tracker: the directory is created in the agent's working
directory, so a repo running agents in subdirectories still ships
`services/api/.claude/` and must add its own anchored entry there.
- **A plain `docker build .` of a clone stamps the version that
`git describe --tags --always` gives**, derived from the `.git` in the build
context as the canonical `Dockerfile` above shows. Without its failure check,
a missing `git` or an unreadable checkout would leave `-X main.Version=` empty
and the build would still exit 0. `script/docker` and `script/cibuild` pass
the version they compute on the host; it takes precedence. They do this
byte-identically across repos:
```sh
# Own line: a failing command substitution inside an argument does not
# trip `set -e`, so the inline form degrades to an empty constant.
version="$(git describe --tags --always --dirty 2>/dev/null || true)"
[ -n "$version" ] || version="unknown"
docker build --no-cache \
--build-arg VERSION="$version" \
-t "$(script/projectname)" .
```
`--always` makes an untagged repo yield an abbreviated commit hash rather
than failing, and the `[ -n "$version" ]` line is the single place the
fallback is applied — a live check that fires on a build from an export with
no `.git` and on a repository with no commits yet. Do not fold it into the
substitution as `|| echo unknown`, which makes the guard unreachable. The
Dockerfile's side is `ARG VERSION` in the stage that compiles, declared
there because `ARG` is stage-scoped; passing `VERSION` to a repo whose
Dockerfile declares no such `ARG` is ignored and costs nothing, which is why
the scripts stay byte-identical. One consequence for CI: the standard
checkout action clones shallow and fetches no tags, so a repo that embeds a
tag-derived version must set `fetch-depth: 0` on its checkout step.
- **Verify `.dockerignore` by enumerating the image, not by reading the
patterns.** Plant files at the root _and_ at least two directories deep, build
a probe image that does `COPY . .`, and list what actually landed
(`docker run --rm --entrypoint find IMAGE /app`). The `transferring context`
size is not a substitute: a nested secret is a few bytes, and BuildKit
transfers only the delta from the previous build.
- **No build artifacts in version control.** Code-derived data (compiled - **No build artifacts in version control.** Code-derived data (compiled
bundles, minified output, generated assets) must never be committed to the bundles, minified output, generated assets) must never be committed to the
@@ -258,9 +476,56 @@ style conventions are in separate documents:
- Make all changes on a feature branch. You can do whatever you want on a - Make all changes on a feature branch. You can do whatever you want on a
feature branch. feature branch.
- `.golangci.yml` is standardized and must _NEVER_ be modified by an agent, only - `.golangci.yml` is standardized. The vendored copy in a consuming repo must
manually by the user. Fetch from _NEVER_ be modified by an agent: fetch it from
`https://git.eeqj.de/sneak/prompts/raw/branch/main/.golangci.yml`. `https://git.eeqj.de/sneak/prompts/raw/branch/main/.golangci.yml` and keep it
byte-identical, so that no repo can quietly loosen its own linting. Linter
configuration changes are made to the canonical copy in the `prompts` repo and
reach consuming repos by re-vendoring; an agent may open a PR against
canonical, which only the user merges. One list is exempt from byte-identity,
because it cannot be written once for every repo: the `deny` list of the
`test-support` depguard rule, where a repo names its own test-support packages
by full import path. A repo adds entries there and changes nothing else, and a
re-vendor carries its entries forward. The canonical golangci-lint version is
v2.14.0 (released 2026-09-24), pinned as the digest of the lint phase's base
image
(`golangci/golangci-lint@sha256:ad862ba6b3798cbe0fd9fd7408d498fd74fbd2623a92406b2fd3898faf0bf98f`,
which reports `2.14.0 built with go1.27.0 from 114493f9`). A module's `go`
directive must not name a newer Go minor version than the one golangci-lint
was built with, or golangci-lint refuses to lint it: this release lints
`go 1.27.1` but not `go 1.28`. That digest is the only pin, since no repo
installs golangci-lint on the host. A repo sets the lint phase digest to the
one named here and re-vendors `.golangci.yml` in the same commit, whichever of
the two prompted the change: the canonical copy can name linters that an older
golangci-lint rejects, and a newer golangci-lint can add linters that
`default: all` switches on until the canonical copy disables them.
- **`script/bootstrap` installs a pinned tool by comparing versions, never by
testing presence.** An `if ! command -v <tool>; then install; fi` guard tests
`PATH` only, so on an already-provisioned machine the pin is inert and a
version bump is a silent no-op — while the Dockerfile, installing into a clean
image, gets the pinned version, so a local `make check` and `make docker` can
disagree about what the tool even is. The canonical form:
- compares the installed version against the pin over the **whole** version
token; a parser that stops at the first `-` reports `2.12.2` for a host
running `2.12.2-rc1` and skips the install;
- treats absent, non-zero, empty or unrecognised `--version` output as a
mismatch, so the failure direction is a redundant install and never a
skipped one;
- after installing, re-resolves the binary the way callers do — `hash -r`,
then through `PATH`, not through the directory the installer wrote to —
and fails naming the resolved path, since an install that a shadowing
binary hides succeeds while changing nothing any caller sees;
- is actually called, and prints the version on both success paths: a
function defined and never invoked has the same exit status and the same
empty output as one that worked.
Keep it POSIX sh: no arrays, no `[[`, no `grep -P`.
A Go tool a repo needs on the host is installed with `go install` pinned to
a commit hash (`go install <package>@<commit hash>`). It is never tracked as
a `go.mod` tool dependency or through a `tools.go` file, either of which
pulls the tool's own dependencies into the repo's `go.mod` and `go.sum`.
- When pinning images or packages by hash, add a comment above the reference - When pinning images or packages by hash, add a comment above the reference
with the version and date (YYYY-MM-DD). with the version and date (YYYY-MM-DD).
@@ -374,12 +639,14 @@ style conventions are in separate documents:
settings. settings.
- Avoid putting files in the repo root unless necessary. Root should contain - Avoid putting files in the repo root unless necessary. Root should contain
only project-level config files (`README.md`, `Makefile`, `Dockerfile`, only project-level config files (`README.md`, `AGENTS.md`, `Makefile`,
`LICENSE`, `.gitignore`, `.editorconfig`, `REPO_POLICIES.md`, and `Dockerfile`, `LICENSE`, `.gitignore`, `.editorconfig`, `REPO_POLICIES.md`,
language-specific config). Everything else goes in a subdirectory. Canonical and language-specific config). Everything else goes in a subdirectory.
subdirectory names: Canonical subdirectory names:
- `bin/` — executable scripts and tools - `bin/` — executable scripts and tools
- `cmd/` — Go command entrypoints - `cmd/` — Go command entrypoints; thin only: one `main.go` per binary whose
body is a single call into `internal/` or `pkg/`, no project logic in
`cmd/`
- `configs/` — configuration templates and examples - `configs/` — configuration templates and examples
- `deploy/` — deployment manifests (k8s, compose, terraform) - `deploy/` — deployment manifests (k8s, compose, terraform)
- `docs/` — documentation and markdown (README.md stays in root) - `docs/` — documentation and markdown (README.md stays in root)
@@ -406,3 +673,7 @@ style conventions are in separate documents:
- Go: `go.mod`, `go.sum`, `.golangci.yml` - Go: `go.mod`, `go.sum`, `.golangci.yml`
- JS: `package.json`, `yarn.lock`, `.prettierrc`, `.prettierignore` - JS: `package.json`, `yarn.lock`, `.prettierrc`, `.prettierignore`
- Python: `pyproject.toml` - Python: `pyproject.toml`
- Guidance for coding agents lives in one `AGENTS.md` at the repository root. It
is never committed under a file or directory named after one agent tool, such
as `CLAUDE.md` or `.claude/`, and never split into separate memory files.
TODO.md
+91 -12
View File
@@ -10,19 +10,100 @@
# Status # Status
pre-1.0. No git tags. Runs in production-style Docker deployment, but pre-1.0. No git tags. The linter and the tests are phases of the Docker
the policy compliance branch (repo-policies-compliance, make check build, and the Gitea workflow runs `script/cibuild` (the format check, both
passing, clean tree) is unmerged to main and the CI workflow is missing. phases, then the image) on every push. The
image sets memory ceilings for a 5 GiB container (README "Memory") and the
README says how to run it under upaas (README "Running under upaas"). A
35-hour run of `3898daa` on the live feed peaked at about 1 GiB, without a
container memory limit.
# Next Step # Next Step
Merge repo-policies-compliance into main (3 commits: policy files and `next` waits for sneak to merge it to `main` through
.gitignore, Makefile targets fmt-check/check/docker/hooks, gofmt pass), https://git.eeqj.de/sneak/routewatch/pulls/6. After that, setting
then add .gitea/workflows/check.yml as a small follow-up commit so CI routewatch up under upaas on fsn1app1 and deploying it are his
runs make check on main. (https://git.eeqj.de/sneak/routewatch/issues/31), and so is the run under a
real 5 GiB limit (https://git.eeqj.de/sneak/routewatch/issues/3).
# Completed Steps # Completed Steps
- 2026-10-14: the shared files are re-vendored from `sneak/prompts` at
`dd4027b`: golangci-lint v2.14.0 and its configuration, lint and test as
separate uncached Dockerfile phases, agent guidance in `AGENTS.md`; the code
is fixed for what the new linter configuration finds (closes #52)
- 2026-10-03: `/api/v1/stats` serves the prefix distribution from memory,
seeded at startup and adjusted on every live-route write, so a request no
longer reads every live route (closes #30)
- 2026-10-03: a new database is created with `auto_vacuum` set to
incremental, through the connection string so it is set before the file
is first written, and each periodic incremental vacuum now returns up to
1000 free pages; the late `PRAGMA auto_vacuum` in `Initialize`, which
SQLite ignored, is gone (closes #43)
- 2026-10-03: looking up an IP address no longer reads every IPv6 route:
both families find the most specific live route with at most 33 or 129
lookups on the prefix index, and the IPv4 range columns are gone. Prefixes
from the feed are stored in one text form, so an IPv6 withdrawal, which the
feed sends uncompressed, now removes its route (closes #48)
- 2026-10-02: a plain `docker build .` stamps the commit's tag or short
commit (`git describe --tags --always`) into the page footer instead of
`unknown`: `.dockerignore` sends `.git` without `.git/config`, a `VERSION`
build argument takes precedence, and `make build` stamps the same value
(closes #46)
- 2026-09-29: the entrypoint creates the data directory if it is missing and
stops the start if a step fails; README "Running under upaas" no longer
asks for the host directory to be created first (closes #42)
- 2026-09-29: `.dockerignore` keeps `.git`, local build output, local
databases and `.env` out of the Docker build context, and so out of the
source archive in the image (closes #39)
- 2026-09-29: README first line names the MIT license and the author;
the License section now says MIT and links `LICENSE`, and an Author
section was added; this file brought up to date (closes #38)
- 2026-09-29: MIT `LICENSE` (closes #1)
- 2026-09-28: stopping the daemon while the feed is flowing no longer
panics with "send on closed channel": the read loop checks for a stop
just before handing a message to the handler queues, and a second
`Stop` no longer closes the queues again (closes #34)
- 2026-09-28: `docker stop` no longer kills the daemon 2 seconds after the
stop signal: the entrypoint switches to the `routewatch` user with
`setpriv` instead of `runuser`, so the daemon receives the signal itself
and gets the whole wait `docker stop` allows, up to its own 60-second
limit (closes #33)
- 2026-09-28: ready to run under upaas: a set but invalid `PORT`,
`XDG_DATA_HOME` or `MALLOC_ARENA_MAX` stops the start, the health
check follows `PORT`, README "Running under upaas" section (closes
#31)
- 2026-09-22: realtime in-memory database statistics: counts seeded at
startup and adjusted on every write, oldest/newest route timestamps via
index-end lookups; `/api/v1/stats` no longer scans the tables (closes
#27)
- 2026-09-21: batch writes take the write lock when their transaction
begins (`_txlock=immediate`), so they wait out a WAL checkpoint instead
of failing with "database is locked" (closes #25)
- 2026-09-21: `MALLOC_ARENA_MAX=2` in the image caps glibc malloc arenas,
so memory outside the Go runtime no longer grows with the core count
(closes #23)
- 2026-09-21: `GOMEMLIMIT=1536MiB` in the image; README Memory section
with the memory budget and the 5 GiB container limit (closes #13)
- 2026-09-21: the four handler queues hold at most 20,000 messages each,
down from 100,000 (closes #11)
- 2026-09-21: two goroutine leaks fixed: the stats handlers after a
timeout and the streamer's tickers on every reconnect (closes #12)
- 2026-09-21: parsed RIS messages no longer keep the unused `Community`
and `Raw` fields (closes #9)
- 2026-09-21: `.editorconfig`, and a Gitea workflow that runs
`script/cibuild` on every push (closes #14)
- 2026-09-21: the peering handler's AS-path map holds at most 500,000
paths and is swapped for an empty one every 30 seconds instead of copied
(closes #10)
- 2026-09-21: SQLite memory bounded across the whole connection pool: a
64 MiB page cache on each connection, 1 GiB soft and 1.5 GiB hard heap
limits (closes #8)
- 2026-09-21: the Docker build runs the format check, the linter and the
tests, linting in a separate stage on a golangci-lint image pinned by
digest (closes #5)
- 2026-09-21: `make test` skips the live-network feed test (`-short`), so
`make check` no longer depends on the network (closes #2)
- 2026-07-07 Adopted scripts-to-rule-them-all: `script/` entrypoints, - 2026-07-07 Adopted scripts-to-rule-them-all: `script/` entrypoints,
Makefile shims, README Entrypoints section Makefile shims, README Entrypoints section
- 2026-02-22: repo policy compliance: required policy files, .gitignore - 2026-02-22: repo policy compliance: required policy files, .gitignore
@@ -41,8 +122,6 @@ runs make check on main.
# Future Steps # Future Steps
- Verify main is green after the merge: make check locally and the new - Production memory under 5 GiB: whether to test under a real 5 GiB
CI workflow passing container limit on fsn1app1 is open for sneak
- Review stale remote branches fix-min-time-calculation and (https://git.eeqj.de/sneak/routewatch/issues/3)
optimize-sqlite-settings: land or delete
- Clean up the tmp/ directory at the repo root: gitignore or remove
cmd/asinfo-gen/main.go
+22 -7
View File
@@ -2,8 +2,10 @@
package main package main
import ( import (
"context"
"encoding/csv" "encoding/csv"
"encoding/json" "encoding/json"
"errors"
"io" "io"
"log" "log"
"net/http" "net/http"
@@ -27,13 +29,17 @@ func main() {
// Fetch the CSV data // Fetch the CSV data
log.Println("Fetching AS CSV data...") log.Println("Fetching AS CSV data...")
resp, err := http.Get(asCsvURL)
req, err := http.NewRequestWithContext(
context.Background(), http.MethodGet, asCsvURL, nil)
if err != nil {
log.Fatalf("Failed to fetch CSV: %v", err)
}
resp, err := http.DefaultClient.Do(req)
if err != nil { if err != nil {
log.Fatalf("Failed to fetch CSV: %v", err) log.Fatalf("Failed to fetch CSV: %v", err)
} }
defer func() {
_ = resp.Body.Close()
}()
if resp.StatusCode != http.StatusOK { if resp.StatusCode != http.StatusOK {
log.Fatalf("Failed to fetch CSV: HTTP %d", resp.StatusCode) log.Fatalf("Failed to fetch CSV: HTTP %d", resp.StatusCode)
@@ -48,17 +54,20 @@ func main() {
log.Fatalf("Failed to read CSV header: %v", err) log.Fatalf("Failed to read CSV header: %v", err)
} }
if len(header) != 3 || header[0] != "asn" || header[1] != "handle" || header[2] != "description" { if len(header) != 3 || header[0] != "asn" || header[1] != "handle" ||
header[2] != "description" {
log.Fatalf("Unexpected CSV header: %v", header) log.Fatalf("Unexpected CSV header: %v", header)
} }
// Read all records // Read all records
var asInfos []ASInfo var asInfos []ASInfo
for { for {
record, err := reader.Read() record, err := reader.Read()
if err == io.EOF { if errors.Is(err, io.EOF) {
break break
} }
if err != nil { if err != nil {
log.Printf("Error reading CSV record: %v", err) log.Printf("Error reading CSV record: %v", err)
@@ -86,12 +95,18 @@ func main() {
}) })
} }
// Closed here rather than deferred: log.Fatalf exits without running
// deferred calls.
_ = resp.Body.Close()
log.Printf("Parsed %d AS records", len(asInfos)) log.Printf("Parsed %d AS records", len(asInfos))
// Convert to JSON and output to stdout // Convert to JSON and output to stdout
encoder := json.NewEncoder(os.Stdout) encoder := json.NewEncoder(os.Stdout)
encoder.SetIndent("", "\t") encoder.SetIndent("", "\t")
if err := encoder.Encode(asInfos); err != nil {
err = encoder.Encode(asInfos)
if err != nil {
log.Fatalf("Failed to encode JSON: %v", err) log.Fatalf("Failed to encode JSON: %v", err)
} }
} }
cmd/streamdumper/main.go
+10 -7
View File
@@ -26,9 +26,17 @@ func main() {
// Register raw message handler that prints to stdout // Register raw message handler that prints to stdout
s.RegisterRawHandler(func(line string) { s.RegisterRawHandler(func(line string) {
fmt.Println(line) _, _ = fmt.Fprintln(os.Stdout, line)
}) })
// Start streaming before any defer: log.Fatal exits without running
// deferred calls.
err := s.Start(context.Background())
if err != nil {
log.Fatal("Failed to start streamer:", err)
}
defer s.Stop()
// Set up context with cancellation // Set up context with cancellation
ctx, cancel := context.WithCancel(context.Background()) ctx, cancel := context.WithCancel(context.Background())
defer cancel() defer cancel()
@@ -36,18 +44,13 @@ func main() {
// Handle signals // Handle signals
sigChan := make(chan os.Signal, 1) sigChan := make(chan os.Signal, 1)
signal.Notify(sigChan, os.Interrupt, syscall.SIGTERM) signal.Notify(sigChan, os.Interrupt, syscall.SIGTERM)
go func() { go func() {
<-sigChan <-sigChan
log.Println("Received shutdown signal") log.Println("Received shutdown signal")
cancel() cancel()
}() }()
// Start streaming
if err := s.Start(); err != nil {
log.Fatal("Failed to start streamer:", err)
}
defer s.Stop()
// Wait for context cancellation // Wait for context cancellation
<-ctx.Done() <-ctx.Done()
} }
entrypoint.sh
+14 -1
View File
@@ -1,7 +1,20 @@
#!/bin/bash #!/bin/bash
set -euo pipefail
# glibc silently ignores a malformed MALLOC_ARENA_MAX, so refuse it here.
if [[ -n "${MALLOC_ARENA_MAX:-}" && ! "$MALLOC_ARENA_MAX" =~ ^[1-9][0-9]*$ ]]; then
echo "MALLOC_ARENA_MAX must be a positive whole number, got '$MALLOC_ARENA_MAX'" >&2
exit 1
fi
# Give the data directory to the routewatch user before the daemon starts,
# whether it is missing, an empty root-owned mount, or holds another uid's files.
mkdir -p /var/lib/berlin.sneak.app.routewatch
cd /var/lib/berlin.sneak.app.routewatch cd /var/lib/berlin.sneak.app.routewatch
chown -R routewatch:routewatch . chown -R routewatch:routewatch .
chmod 700 . chmod 700 .
exec runuser -u routewatch -- /app/routewatch # setpriv replaces itself with the daemon, so the daemon receives the stop
# signal directly. runuser would stay in between and kill the daemon 2 seconds
# after passing the signal on.
exec setpriv --reuid=routewatch --regid=routewatch --init-groups -- /app/routewatch
internal/config/config.go
+64 -10
View File
@@ -2,10 +2,12 @@
package config package config
import ( import (
"errors"
"fmt" "fmt"
"os" "os"
"path/filepath" "path/filepath"
"runtime" "runtime"
"strconv"
"time" "time"
) )
@@ -18,6 +20,18 @@ const (
// defaultRouteExpirationMinutes is the default route expiration timeout in minutes // defaultRouteExpirationMinutes is the default route expiration timeout in minutes
defaultRouteExpirationMinutes = 5 defaultRouteExpirationMinutes = 5
// defaultPort is the HTTP port used when PORT is not set
defaultPort = 8080
// maxPort is the highest TCP port number
maxPort = 65535
)
var (
errRelativeXDGDataHome = errors.New("XDG_DATA_HOME must be an absolute path")
errUnsupportedOS = errors.New("unsupported operating system")
errInvalidPort = errors.New("PORT must be a whole number")
) )
// Config holds configuration for the entire application // Config holds configuration for the entire application
@@ -25,14 +39,20 @@ type Config struct {
// StateDir is the directory for all application state (database, snapshots) // StateDir is the directory for all application state (database, snapshots)
StateDir string StateDir string
// Port is the TCP port the HTTP server listens on
Port int
// MaxRuntime is the maximum runtime (0 = run forever) // MaxRuntime is the maximum runtime (0 = run forever)
MaxRuntime time.Duration MaxRuntime time.Duration
// EnableBatchedDatabaseWrites enables batched database operations for better performance // EnableBatchedDatabaseWrites enables batched database operations for
// better performance
EnableBatchedDatabaseWrites bool EnableBatchedDatabaseWrites bool
// RouteExpirationTimeout is how long a route can go without being refreshed before expiring // RouteExpirationTimeout is how long a route can go without being
// Default is 2 hours which is conservative for BGP (typical BGP hold time is 90-180 seconds) // refreshed before expiring
// Default is 2 hours which is conservative for BGP (typical BGP hold
// time is 90-180 seconds)
RouteExpirationTimeout time.Duration RouteExpirationTimeout time.Duration
} }
@@ -43,11 +63,20 @@ func New() (*Config, error) {
return nil, fmt.Errorf("failed to determine state directory: %w", err) return nil, fmt.Errorf("failed to determine state directory: %w", err)
} }
port, err := getPort()
if err != nil {
return nil, err
}
return &Config{ return &Config{
StateDir: stateDir, StateDir: stateDir,
MaxRuntime: 0, // Run forever by default Port: port,
EnableBatchedDatabaseWrites: true, // Enable batching by default // Run forever by default
RouteExpirationTimeout: defaultRouteExpirationMinutes * time.Minute, // For active route monitoring MaxRuntime: 0,
// Enable batching by default
EnableBatchedDatabaseWrites: true,
// For active route monitoring
RouteExpirationTimeout: defaultRouteExpirationMinutes * time.Minute,
}, nil }, nil
} }
@@ -69,13 +98,20 @@ func getStateDirectory() (string, error) {
return filepath.Join(home, "Library", "Application Support", AppIdentifier), nil return filepath.Join(home, "Library", "Application Support", AppIdentifier), nil
case "linux", "freebsd", "openbsd", "netbsd": case "linux", "freebsd", "openbsd", "netbsd":
// The XDG spec requires an absolute path; a relative one would put
// the database somewhere unexpected.
xdgData := os.Getenv("XDG_DATA_HOME")
if xdgData != "" && !filepath.IsAbs(xdgData) {
return "", fmt.Errorf("%w, got %q", errRelativeXDGDataHome, xdgData)
}
// Unix-like: /var/lib/berlin.sneak.app.routewatch if root, else XDG_DATA_HOME // Unix-like: /var/lib/berlin.sneak.app.routewatch if root, else XDG_DATA_HOME
if os.Geteuid() == 0 { if os.Geteuid() == 0 {
return filepath.Join("/var/lib", AppIdentifier), nil return filepath.Join("/var/lib", AppIdentifier), nil
} }
// Check XDG_DATA_HOME first // Check XDG_DATA_HOME first
if xdgData := os.Getenv("XDG_DATA_HOME"); xdgData != "" { if xdgData != "" {
return filepath.Join(xdgData, AppIdentifier), nil return filepath.Join(xdgData, AppIdentifier), nil
} }
@@ -88,14 +124,32 @@ func getStateDirectory() (string, error) {
return filepath.Join(home, ".local", "share", AppIdentifier), nil return filepath.Join(home, ".local", "share", AppIdentifier), nil
default: default:
return "", fmt.Errorf("unsupported operating system: %s", runtime.GOOS) return "", fmt.Errorf("%w: %s", errUnsupportedOS, runtime.GOOS)
} }
} }
// getPort returns the HTTP port from PORT, or defaultPort when PORT is not set
func getPort() (int, error) {
value := os.Getenv("PORT")
if value == "" {
return defaultPort, nil
}
// ParseUint, unlike Atoi, refuses a sign: the health check URL cannot use "+9090"
port, err := strconv.ParseUint(value, 10, 0)
if err != nil || port < 1 || port > maxPort {
return 0, fmt.Errorf("%w from 1 to %d, got %q",
errInvalidPort, maxPort, value)
}
return int(port), nil
}
// EnsureDirectories creates all necessary directories if they don't exist // EnsureDirectories creates all necessary directories if they don't exist
func (c *Config) EnsureDirectories() error { func (c *Config) EnsureDirectories() error {
// Ensure state directory exists // Ensure state directory exists
if err := os.MkdirAll(c.StateDir, dirPermissions); err != nil { err := os.MkdirAll(c.StateDir, dirPermissions)
if err != nil {
return fmt.Errorf("failed to create state directory: %w", err) return fmt.Errorf("failed to create state directory: %w", err)
} }
internal/config/config_internal_test.go
+67
View File
@@ -0,0 +1,67 @@
package config
import (
"runtime"
"testing"
)
func TestNewReadsPort(t *testing.T) {
tests := map[string]int{
"": defaultPort,
"1": 1,
"9090": 9090,
"65535": 65535,
}
for value, want := range tests {
t.Run(value, func(t *testing.T) {
t.Setenv("PORT", value)
t.Setenv("XDG_DATA_HOME", "")
cfg, err := New()
if err != nil {
t.Fatalf("New() with PORT=%q: %v", value, err)
}
if cfg.Port != want {
t.Errorf("New() with PORT=%q: Port = %d, want %d", value, cfg.Port, want)
}
})
}
}
func TestNewRefusesInvalidPort(t *testing.T) {
for _, value := range []string{"0", "65536", "-1", "+9090", "http", "80.5"} {
t.Run(value, func(t *testing.T) {
t.Setenv("PORT", value)
t.Setenv("XDG_DATA_HOME", "")
_, err := New()
if err == nil {
t.Errorf("New() with PORT=%q returned no error", value)
}
})
}
}
func TestNewRefusesRelativeXDGDataHome(t *testing.T) {
if runtime.GOOS == "darwin" {
t.Skip("macOS does not read XDG_DATA_HOME")
}
t.Setenv("PORT", "")
t.Setenv("XDG_DATA_HOME", "relative/path")
_, err := New()
if err == nil {
t.Error("New() with a relative XDG_DATA_HOME returned no error")
}
t.Setenv("XDG_DATA_HOME", "/var/lib")
_, err = New()
if err != nil {
t.Errorf("New() with XDG_DATA_HOME=/var/lib: %v", err)
}
}
internal/database/counts.go
+218
View File
@@ -0,0 +1,218 @@
package database
import (
"context"
"fmt"
"sync"
)
// liveCounts holds the running row counts and the prefix distribution that the
// stats endpoints report. They are seeded once at startup from the tables and
// then adjusted on every write, so a stats read serves them from memory instead
// of running a query over the tables. The COUNT(*) scans (issue 27) and then
// the prefix distribution query (issue 30) each grew with the database until
// they took the whole request timeout and made /api/v1/stats return 500.
//
// A single mutex guards all fields so the stats reader takes a consistent
// snapshot at one instant and writers, which already run under the database
// write lock, adjust the counts after their transaction commits.
type liveCounts struct {
mu sync.RWMutex
asns int
prefixesV4 int
prefixesV6 int
peerings int
peers int
routesV4 int
routesV6 int
// The prefix distribution: for each mask length, the number of distinct
// prefixes that have at least one live route.
distributionV4 [ipv4Bits + 1]int
distributionV6 [ipv6Bits + 1]int
}
// seed sets every count to the value read from the tables at startup. It runs
// before any writer, so it needs no coordination with the adjust methods.
func (c *liveCounts) seed(
asns, prefixesV4, prefixesV6, peerings, peers, routesV4, routesV6 int,
distributionV4, distributionV6 []PrefixDistribution,
) {
c.mu.Lock()
defer c.mu.Unlock()
c.asns = asns
c.prefixesV4 = prefixesV4
c.prefixesV6 = prefixesV6
c.peerings = peerings
c.peers = peers
c.routesV4 = routesV4
c.routesV6 = routesV6
for _, entry := range distributionV4 {
addAtMaskLength(c.distributionV4[:], entry.MaskLength, entry.Count)
}
for _, entry := range distributionV6 {
addAtMaskLength(c.distributionV6[:], entry.MaskLength, entry.Count)
}
}
// addASNs adds n to the ASN count.
func (c *liveCounts) addASNs(n int) {
c.mu.Lock()
c.asns += n
c.mu.Unlock()
}
// addPrefixes adds to the IPv4 and IPv6 prefix counts.
func (c *liveCounts) addPrefixes(v4, v6 int) {
c.mu.Lock()
c.prefixesV4 += v4
c.prefixesV6 += v6
c.mu.Unlock()
}
// addPeerings adds n to the peering count.
func (c *liveCounts) addPeerings(n int) {
c.mu.Lock()
c.peerings += n
c.mu.Unlock()
}
// addPeers adds n to the BGP peer count.
func (c *liveCounts) addPeers(n int) {
c.mu.Lock()
c.peers += n
c.mu.Unlock()
}
// addRoutes adds to the IPv4 and IPv6 live-route counts. Deletions pass
// negative values.
func (c *liveCounts) addRoutes(v4, v6 int) {
c.mu.Lock()
c.routesV4 += v4
c.routesV6 += v6
c.mu.Unlock()
}
// addToDistribution adds n to the IPv4 and IPv6 prefix distributions once for
// each listed mask length. A write lists the mask lengths of the prefixes it
// gave their first live route with n = 1, and of the prefixes it left with no
// live route with n = -1.
func (c *liveCounts) addToDistribution(maskLengthsV4, maskLengthsV6 []int, n int) {
c.mu.Lock()
defer c.mu.Unlock()
for _, maskLength := range maskLengthsV4 {
addAtMaskLength(c.distributionV4[:], maskLength, n)
}
for _, maskLength := range maskLengthsV6 {
addAtMaskLength(c.distributionV6[:], maskLength, n)
}
}
// addAtMaskLength adds n to counts[maskLength]. A mask length the array has no
// entry for is ignored, so a malformed route cannot crash the daemon.
func addAtMaskLength(counts []int, maskLength, n int) {
if maskLength >= 0 && maskLength < len(counts) {
counts[maskLength] += n
}
}
// fill copies the counts into a Stats, including the derived totals, under a
// single read lock so the reader sees one consistent snapshot.
func (c *liveCounts) fill(s *Stats) {
c.mu.RLock()
defer c.mu.RUnlock()
s.ASNs = c.asns
s.IPv4Prefixes = c.prefixesV4
s.IPv6Prefixes = c.prefixesV6
s.Prefixes = c.prefixesV4 + c.prefixesV6
s.Peerings = c.peerings
s.Peers = c.peers
s.IPv4Routes = c.routesV4
s.IPv6Routes = c.routesV6
s.LiveRoutes = c.routesV4 + c.routesV6
s.IPv4PrefixDistribution = distributionList(c.distributionV4[:])
s.IPv6PrefixDistribution = distributionList(c.distributionV6[:])
}
// distributionList lists the mask lengths that have at least one prefix, in
// ascending order, the way the distribution query returns them.
func distributionList(counts []int) []PrefixDistribution {
var list []PrefixDistribution
for maskLength, count := range counts {
if count > 0 {
list = append(list,
PrefixDistribution{MaskLength: maskLength, Count: count})
}
}
return list
}
// countRows returns the number of rows in the named table. It is used only at
// startup to seed the in-memory counters, so a full COUNT(*) is acceptable.
func (d *Database) countRows(ctx context.Context, table string) (int, error) {
var n int
// table is one of a fixed set of literals below, never external input.
err := d.db.QueryRowContext(ctx, "SELECT COUNT(*) FROM "+table).Scan(&n)
if err != nil {
return 0, fmt.Errorf("failed to count %s: %w", table, err)
}
return n, nil
}
// seedCounts reads the current row counts and prefix distribution from the
// tables into the in-memory counters. It runs once at startup, before the
// streamer begins writing.
func (d *Database) seedCounts(ctx context.Context) error {
asns, err := d.countRows(ctx, "asns")
if err != nil {
return err
}
prefixesV4, err := d.countRows(ctx, "prefixes_v4")
if err != nil {
return err
}
prefixesV6, err := d.countRows(ctx, "prefixes_v6")
if err != nil {
return err
}
peerings, err := d.countRows(ctx, "peerings")
if err != nil {
return err
}
peers, err := d.countRows(ctx, "bgp_peers")
if err != nil {
return err
}
routesV4, err := d.countRows(ctx, tableLiveRoutesV4)
if err != nil {
return err
}
routesV6, err := d.countRows(ctx, tableLiveRoutesV6)
if err != nil {
return err
}
distributionV4, distributionV6, err := d.GetPrefixDistributionContext(ctx)
if err != nil {
return err
}
d.counts.seed(asns, prefixesV4, prefixesV6, peerings, peers, routesV4, routesV6,
distributionV4, distributionV6)
return nil
}
internal/database/counts_internal_test.go
+614
View File
@@ -0,0 +1,614 @@
package database
import (
"context"
"slices"
"sync"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/logger"
"github.com/google/uuid"
)
// testV4PeerIP is the peer the IPv4 routes in these tests are learned from.
const testV4PeerIP = "192.0.2.1"
// testV6Prefix is the prefix of every IPv6 route in these tests.
const testV6Prefix = "2001:db8::/32"
// mkV4Route builds an IPv4 live route with its mask length taken from the
// prefix.
func mkV4Route(t *testing.T, prefix string, asn int, ts time.Time) *LiveRoute {
t.Helper()
maskLength, err := prefixMaskLength(prefix)
if err != nil {
t.Fatalf("prefixMaskLength(%s): %v", prefix, err)
}
return &LiveRoute{
ID: uuid.New(),
Prefix: prefix,
MaskLength: maskLength,
IPVersion: ipVersionV4,
OriginASN: asn,
PeerIP: testV4PeerIP,
ASPath: []int{asn},
NextHop: "192.0.2.254",
LastUpdated: ts,
}
}
// mkV6Route builds an IPv6 live route for testV6Prefix.
func mkV6Route(asn int, ts time.Time) *LiveRoute {
return &LiveRoute{
ID: uuid.New(),
Prefix: testV6Prefix,
MaskLength: 32,
IPVersion: ipVersionV6,
OriginASN: asn,
PeerIP: "2001:db8::1",
ASPath: []int{asn},
NextHop: "2001:db8::ffff",
LastUpdated: ts,
}
}
// TestLiveCountsTrackWritesInRealtime checks that the stats counts start at
// zero, reflect each write the moment it commits (no recompute, no timer), do
// not move when a route is merely re-announced, and drop when a route is
// deleted. These counts are what /api/v1/stats reports; before this change the
// endpoint recomputed them with a COUNT(*) over each table on every request.
func TestLiveCountsTrackWritesInRealtime(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx := context.Background()
empty, err := db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext on empty database: %v", err)
}
if empty.ASNs != 0 || empty.Prefixes != 0 || empty.Peerings != 0 ||
empty.Peers != 0 || empty.LiveRoutes != 0 {
t.Fatalf("empty database counts nonzero: %+v", empty)
}
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
writeASNsPrefixesAndPeers(t, db, ts)
routes := []*LiveRoute{
mkV4Route(t, "198.51.100.0/24", 64500, ts),
mkV4Route(t, "203.0.113.0/24", 64501, ts.Add(time.Minute)),
mkV6Route(64502, ts.Add(2*time.Minute)),
}
err = db.UpsertLiveRouteBatch(routes)
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
stats, err := db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext: %v", err)
}
assertCounts(t, "after inserts", stats, wantCounts{
asns: 2, prefixes: 2, peerings: 1, peers: 1,
ipv4Routes: 2, ipv6Routes: 1, liveRoutes: 3,
})
// Re-announcing the same routes is an update, not an insert: counts hold.
err = db.UpsertLiveRouteBatch(routes)
if err != nil {
t.Fatalf("UpsertLiveRouteBatch (re-announce): %v", err)
}
stats, err = db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext: %v", err)
}
assertCounts(t, "after re-announce", stats, wantCounts{
asns: 2, prefixes: 2, peerings: 1, peers: 1,
ipv4Routes: 2, ipv6Routes: 1, liveRoutes: 3,
})
// A withdrawal removes one route.
err = db.DeleteLiveRouteBatch([]LiveRouteDeletion{{
Prefix: "203.0.113.0/24", OriginASN: 64501,
PeerIP: testV4PeerIP, IPVersion: ipVersionV4,
}})
if err != nil {
t.Fatalf("DeleteLiveRouteBatch: %v", err)
}
stats, err = db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext: %v", err)
}
assertCounts(t, "after delete", stats, wantCounts{
asns: 2, prefixes: 2, peerings: 1, peers: 1,
ipv4Routes: 1, ipv6Routes: 1, liveRoutes: 2,
})
}
// writeASNsPrefixesAndPeers writes two ASNs, two prefixes, one peer and one
// peering: the rows besides live routes that the stats count.
func writeASNsPrefixesAndPeers(t *testing.T, db *Database, ts time.Time) {
t.Helper()
err := db.GetOrCreateASNBatch(map[int]time.Time{64500: ts, 64501: ts})
if err != nil {
t.Fatalf("GetOrCreateASNBatch: %v", err)
}
err = db.UpdatePrefixesBatch(map[string]time.Time{
"198.51.100.0/24": ts,
testV6Prefix: ts,
})
if err != nil {
t.Fatalf("UpdatePrefixesBatch: %v", err)
}
err = db.UpdatePeerBatch(map[string]PeerUpdate{
testV4PeerIP: {
PeerIP: testV4PeerIP, PeerASN: 64500,
MessageType: "UPDATE", Timestamp: ts,
},
})
if err != nil {
t.Fatalf("UpdatePeerBatch: %v", err)
}
err = db.RecordPeering(64500, 64501, ts)
if err != nil {
t.Fatalf("RecordPeering: %v", err)
}
}
// TestLiveCountsSeededFromDatabaseAtStartup writes rows, reopens the same
// database file, and checks the counts and the prefix distribution come back
// from the seed scan rather than starting at zero.
func TestLiveCountsSeededFromDatabaseAtStartup(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
err = db.GetOrCreateASNBatch(map[int]time.Time{64500: ts, 64501: ts, 64502: ts})
if err != nil {
t.Fatalf("GetOrCreateASNBatch: %v", err)
}
err = db.UpsertLiveRouteBatch([]*LiveRoute{
mkV4Route(t, "198.51.100.0/24", 64500, ts),
mkV6Route(64502, ts),
})
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
err = db.Close()
if err != nil {
t.Fatalf("Close: %v", err)
}
reopened, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to reopen database: %v", err)
}
defer func() { _ = reopened.Close() }()
stats, err := reopened.GetStatsContext(context.Background())
if err != nil {
t.Fatalf("GetStatsContext after reopen: %v", err)
}
if stats.ASNs != 3 {
t.Errorf("seeded ASNs = %d, want 3", stats.ASNs)
}
if stats.IPv4Routes != 1 || stats.IPv6Routes != 1 || stats.LiveRoutes != 2 {
t.Errorf("seeded routes = (v4 %d, v6 %d, total %d), want (1, 1, 2)",
stats.IPv4Routes, stats.IPv6Routes, stats.LiveRoutes)
}
assertDistribution(t, "seeded IPv4 distribution", stats.IPv4PrefixDistribution,
[]PrefixDistribution{{MaskLength: 24, Count: 1}})
assertDistribution(t, "seeded IPv6 distribution", stats.IPv6PrefixDistribution,
[]PrefixDistribution{{MaskLength: 32, Count: 1}})
}
// TestPrefixDistributionTracksWrites checks that the prefix distribution the
// stats read reports stays exact across each kind of live-route write, and that
// after every step it equals what the distribution query reads from the route
// tables. The steps run once through the batch methods the prefix handler uses
// and once through the single-route methods.
func TestPrefixDistributionTracksWrites(t *testing.T) {
t.Parallel()
steps := prefixDistributionSteps(t)
for _, batch := range []bool{true, false} {
name := "single-route writes"
if batch {
name = "batch writes"
}
t.Run(name, func(t *testing.T) {
t.Parallel()
db, err := New(&config.Config{StateDir: t.TempDir()}, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx := context.Background()
for _, step := range steps {
err := announceRoutes(db, batch, step.announce)
if err != nil {
t.Fatalf("%s: announce: %v", step.name, err)
}
err = withdrawRoutes(db, batch, step.withdraw)
if err != nil {
t.Fatalf("%s: withdraw: %v", step.name, err)
}
stats, err := db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("%s: GetStatsContext: %v", step.name, err)
}
queryV4, queryV6, err := db.GetPrefixDistributionContext(ctx)
if err != nil {
t.Fatalf("%s: GetPrefixDistributionContext: %v", step.name, err)
}
assertDistribution(t, step.name+": IPv4 distribution",
stats.IPv4PrefixDistribution, step.wantV4)
assertDistribution(t, step.name+": IPv6 distribution",
stats.IPv6PrefixDistribution, step.wantV6)
assertDistribution(t, step.name+": IPv4 distribution query",
queryV4, step.wantV4)
assertDistribution(t, step.name+": IPv6 distribution query",
queryV6, step.wantV6)
}
})
}
}
// prefixDistributionStep is one step of TestPrefixDistributionTracksWrites:
// the routes it announces, then the routes it withdraws, then the
// distributions both the stats and the distribution query must report.
type prefixDistributionStep struct {
name string
announce []*LiveRoute
withdraw []*LiveRoute
wantV4 []PrefixDistribution
wantV6 []PrefixDistribution
}
// prefixDistributionSteps returns the steps of
// TestPrefixDistributionTracksWrites, in the order they run.
func prefixDistributionSteps(t *testing.T) []prefixDistributionStep {
t.Helper()
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
shared := mkV4Route(t, "198.51.100.0/24", 64500, ts)
sharedSecondPeer := mkV4Route(t, "198.51.100.0/24", 64500, ts)
sharedSecondPeer.PeerIP = "192.0.2.2"
other := mkV4Route(t, "203.0.113.0/24", 64501, ts)
wide := mkV4Route(t, "172.16.0.0/16", 64502, ts)
v6 := mkV6Route(64503, ts)
v6SecondPeer := mkV6Route(64503, ts)
v6SecondPeer.PeerIP = "2001:db8::2"
all := []PrefixDistribution{{MaskLength: 16, Count: 1}, {MaskLength: 24, Count: 2}}
v6Only := []PrefixDistribution{{MaskLength: 32, Count: 1}}
return []prefixDistributionStep{
{
name: "new routes",
announce: []*LiveRoute{shared, other, wide, v6},
wantV4: all, wantV6: v6Only,
},
{
name: "re-announcement",
announce: []*LiveRoute{shared, other, wide, v6},
wantV4: all, wantV6: v6Only,
},
{
name: "second peer announces a prefix that has a live route",
announce: []*LiveRoute{sharedSecondPeer, v6SecondPeer},
wantV4: all, wantV6: v6Only,
},
{
name: "withdrawal of a route that is not the last for its prefix",
withdraw: []*LiveRoute{shared, v6},
wantV4: all, wantV6: v6Only,
},
{
name: "withdrawal of the last route for a prefix",
withdraw: []*LiveRoute{sharedSecondPeer, v6SecondPeer},
wantV4: []PrefixDistribution{
{MaskLength: 16, Count: 1}, {MaskLength: 24, Count: 1},
},
},
{
name: "withdrawal of every remaining route",
withdraw: []*LiveRoute{other, wide},
},
{
name: "two peers announce a new prefix together",
announce: []*LiveRoute{shared, sharedSecondPeer, v6, v6SecondPeer},
wantV4: []PrefixDistribution{{MaskLength: 24, Count: 1}},
wantV6: v6Only,
},
{
name: "both routes for a prefix withdrawn together",
withdraw: []*LiveRoute{shared, sharedSecondPeer, v6, v6SecondPeer},
},
// The feed often withdraws a route that is not live. That must not take
// the prefix out of the distribution. A count wrongly taken below zero is
// left out of the answer, so the next step shows it: its announcement
// at the same mask length would then not be counted.
{
name: "withdrawal of routes that are not live",
withdraw: []*LiveRoute{other, v6},
},
{
name: "announcement after a withdrawal of routes that are not live",
announce: []*LiveRoute{other, v6},
wantV4: []PrefixDistribution{{MaskLength: 24, Count: 1}},
wantV6: v6Only,
},
}
}
// announceRoutes writes routes in one UpsertLiveRouteBatch, or with one
// UpsertLiveRoute each.
func announceRoutes(db *Database, batch bool, routes []*LiveRoute) error {
if batch {
return db.UpsertLiveRouteBatch(routes)
}
for _, route := range routes {
err := db.UpsertLiveRoute(route)
if err != nil {
return err
}
}
return nil
}
// withdrawRoutes removes routes in one DeleteLiveRouteBatch, or with one
// DeleteLiveRoute each. It names each route by prefix and peer only, with no
// origin ASN, as a withdrawal from the feed does when its message carries no AS
// path.
func withdrawRoutes(db *Database, batch bool, routes []*LiveRoute) error {
if !batch {
for _, route := range routes {
err := db.DeleteLiveRoute(route.Prefix, 0, route.PeerIP)
if err != nil {
return err
}
}
return nil
}
deletions := make([]LiveRouteDeletion, 0, len(routes))
for _, route := range routes {
deletions = append(deletions, LiveRouteDeletion{
Prefix: route.Prefix, PeerIP: route.PeerIP, IPVersion: route.IPVersion,
})
}
return db.DeleteLiveRouteBatch(deletions)
}
// TestStatsRouteTimestamps checks the oldest/newest route timestamps are read
// from the right rows across both tables and parse into time.Time. The old
// MIN/MAX union query read its result into *time.Time, which the driver could
// not parse, so it logged a warning every call and left both timestamps nil.
func TestStatsRouteTimestamps(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx := context.Background()
empty, err := db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext on empty database: %v", err)
}
if empty.OldestRoute != nil || empty.NewestRoute != nil {
t.Fatalf("empty database timestamps = (%v, %v), want (nil, nil)",
empty.OldestRoute, empty.NewestRoute)
}
base := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
oldest := base
newest := base.Add(2 * time.Minute)
err = db.UpsertLiveRouteBatch([]*LiveRoute{
mkV4Route(t, "198.51.100.0/24", 64500, base.Add(time.Minute)),
mkV4Route(t, "203.0.113.0/24", 64501, oldest),
mkV6Route(64502, newest),
})
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
stats, err := db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext: %v", err)
}
if stats.OldestRoute == nil || !stats.OldestRoute.Equal(oldest) {
t.Errorf("OldestRoute = %v, want %v", stats.OldestRoute, oldest)
}
if stats.NewestRoute == nil || !stats.NewestRoute.Equal(newest) {
t.Errorf("NewestRoute = %v, want %v", stats.NewestRoute, newest)
}
}
// TestLiveCountsConcurrentReadWrite runs writers and stats readers at once so
// the race detector proves the counters are safe under concurrent use.
func TestLiveCountsConcurrentReadWrite(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
const writers = 4
var wg sync.WaitGroup
wg.Add(writers)
for w := range writers {
go func(base int) {
defer wg.Done()
for i := range 25 {
asn := 65000 + base*100 + i
route := mkV6Route(asn, ts)
route.PeerIP = "2001:db8::" + uuid.NewString()
err := db.UpsertLiveRoute(route)
if err != nil {
t.Errorf("UpsertLiveRoute: %v", err)
return
}
}
}(w)
}
var readerWG sync.WaitGroup
readerWG.Add(1)
stop := make(chan struct{})
go func() {
defer readerWG.Done()
for {
select {
case <-stop:
return
default:
_, err := db.GetStatsContext(context.Background())
if err != nil {
t.Errorf("GetStatsContext: %v", err)
return
}
}
}
}()
wg.Wait()
close(stop)
readerWG.Wait()
stats, err := db.GetStatsContext(context.Background())
if err != nil {
t.Fatalf("GetStatsContext: %v", err)
}
if want := writers * 25; stats.IPv6Routes != want {
t.Errorf("IPv6Routes = %d, want %d", stats.IPv6Routes, want)
}
// Every writer announced the same prefix, so it counts once.
assertDistribution(t, "IPv6 distribution", stats.IPv6PrefixDistribution,
[]PrefixDistribution{{MaskLength: 32, Count: 1}})
}
type wantCounts struct {
asns int
prefixes int
peerings int
peers int
ipv4Routes int
ipv6Routes int
liveRoutes int
}
func assertCounts(t *testing.T, when string, got Stats, want wantCounts) {
t.Helper()
if got.ASNs != want.asns {
t.Errorf("%s: ASNs = %d, want %d", when, got.ASNs, want.asns)
}
if got.Prefixes != want.prefixes {
t.Errorf("%s: Prefixes = %d, want %d", when, got.Prefixes, want.prefixes)
}
if got.Peerings != want.peerings {
t.Errorf("%s: Peerings = %d, want %d", when, got.Peerings, want.peerings)
}
if got.Peers != want.peers {
t.Errorf("%s: Peers = %d, want %d", when, got.Peers, want.peers)
}
if got.IPv4Routes != want.ipv4Routes {
t.Errorf("%s: IPv4Routes = %d, want %d", when, got.IPv4Routes, want.ipv4Routes)
}
if got.IPv6Routes != want.ipv6Routes {
t.Errorf("%s: IPv6Routes = %d, want %d", when, got.IPv6Routes, want.ipv6Routes)
}
if got.LiveRoutes != want.liveRoutes {
t.Errorf("%s: LiveRoutes = %d, want %d", when, got.LiveRoutes, want.liveRoutes)
}
}
func assertDistribution(t *testing.T, what string, got, want []PrefixDistribution) {
t.Helper()
if !slices.Equal(got, want) {
t.Errorf("%s = %v, want %v", what, got, want)
}
}
internal/database/database.go
+1366 -1050
View File
File diff suppressed because it is too large Load Diff
internal/database/database_internal_test.go
+434
View File
@@ -0,0 +1,434 @@
package database
import (
"context"
"database/sql"
"errors"
"fmt"
"net/netip"
"sync"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/logger"
"github.com/google/uuid"
)
// tempStoreMemory is the PRAGMA temp_store value meaning "hold temp B-trees in
// memory"; the DSN change must leave temp_store below this so they spill to disk.
const tempStoreMemory = 2
// autoVacuumIncremental is the PRAGMA auto_vacuum value meaning "incremental".
const autoVacuumIncremental = 2
// vacuumTestRoutes is how many routes the vacuum test writes and then deletes,
// enough to leave many free pages in the file.
const vacuumTestRoutes = 2000
// heldConnections is how many pooled connections the pragma test holds open at
// once so each is a distinct SQLite connection that parsed the DSN.
const heldConnections = 5
// testPeerIP is the peer every route in the IP lookup test is learned from.
const testPeerIP = "192.0.2.254"
// Parameters for the checkpoint-contention regression test.
const (
// contentionIterations is how many batch writes race the checkpoint loop.
contentionIterations = 400
// contendedASNCount is the small set of ASNs the batches reuse, so most
// batches update existing rows and exercise the read-before-write path.
contendedASNCount = 16
// asnSecondBand offsets a second ASN per batch so each batch writes more
// than one row.
asnSecondBand = 100
)
// TestGetIPInfoFindsMostSpecificLiveRoute stores nested live prefixes for both
// families and checks that a lookup returns the most specific one covering the
// address, ErrNoRoute when none covers it, and the next less specific prefix
// once the only route of the most specific one is withdrawn.
func TestGetIPInfoFindsMostSpecificLiveRoute(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
// Nested live prefixes, each originated by its own AS.
origins := map[string]int{
"10.0.0.0/8": 64500,
"10.1.0.0/16": 64501,
"10.1.2.0/24": 64502,
"2001:db8::/32": 64500,
"2001:db8:1::/48": 64501,
"2001:db8:1:2::/64": 64502,
}
writeOriginRoutes(t, db, origins)
// lookup checks that ip resolves to the live prefix want, or to ErrNoRoute
// when want is empty.
lookup := func(ip, want string) {
t.Helper()
info, err := db.GetIPInfo(ip)
if want == "" {
if !errors.Is(err, ErrNoRoute) {
t.Errorf("GetIPInfo(%s) = %+v, %v; want ErrNoRoute", ip, info, err)
}
return
}
if err != nil {
t.Errorf("GetIPInfo(%s): %v", ip, err)
return
}
if info.Netblock != want || info.MaskLength != netip.MustParsePrefix(want).Bits() ||
info.ASN != origins[want] {
t.Errorf("GetIPInfo(%s) = %s (mask %d) AS%d, want %s AS%d",
ip, info.Netblock, info.MaskLength, info.ASN, want, origins[want])
}
}
lookup("10.1.2.3", "10.1.2.0/24")
lookup("::ffff:10.1.2.3", "10.1.2.0/24")
lookup("10.1.3.4", "10.1.0.0/16")
lookup("10.2.0.1", "10.0.0.0/8")
lookup("192.0.2.1", "")
lookup("2001:db8:1:2::3", "2001:db8:1:2::/64")
lookup("2001:db8:1:3::4", "2001:db8:1::/48")
lookup("2001:db8:2::1", "2001:db8::/32")
lookup("2001:db9::1", "")
err = db.DeleteLiveRouteBatch([]LiveRouteDeletion{
{
Prefix: "10.1.2.0/24", OriginASN: 64502,
PeerIP: testPeerIP, IPVersion: ipVersionV4,
},
{
Prefix: "2001:db8:1:2::/64", OriginASN: 64502,
PeerIP: testPeerIP, IPVersion: ipVersionV6,
},
})
if err != nil {
t.Fatalf("DeleteLiveRouteBatch: %v", err)
}
lookup("10.1.2.3", "10.1.0.0/16")
lookup("2001:db8:1:2::3", "2001:db8:1::/48")
}
// TestGetIPInfoIgnoresShortRoutes checks that the IP lookup does not report a
// default route, or a route shorter than /8 for IPv4 or /16 for IPv6, as the
// route for an address, and that a route of exactly /8 or /16, or longer, still
// is. The prefixes are written out so that a wrong limit fails the test.
func TestGetIPInfoIgnoresShortRoutes(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
// announce writes one live route for prefix, originated by asn.
announce := func(prefix string, asn int) {
t.Helper()
writeOriginRoutes(t, db, map[string]int{prefix: asn})
}
// lookup checks that ip resolves to the live prefix netblock from asn.
lookup := func(ip, netblock string, asn int) {
t.Helper()
info, err := db.GetIPInfo(ip)
if err != nil {
t.Errorf("GetIPInfo(%s): %v", ip, err)
return
}
if info.Netblock != netblock || info.ASN != asn {
t.Errorf("GetIPInfo(%s) = %s AS%d, want %s AS%d",
ip, info.Netblock, info.ASN, netblock, asn)
}
}
// Default and very short routes, and routes one bit shorter than each limit.
announce("2000::/3", 64500)
announce("0.0.0.0/0", 64500)
announce("2000::/15", 64500)
announce("192.0.0.0/7", 64500)
for _, ip := range []string{"2001:db8::1", "192.0.2.1"} {
info, err := db.GetIPInfo(ip)
if !errors.Is(err, ErrNoRoute) {
t.Errorf("GetIPInfo(%s) with only short routes = %+v, %v; "+
"want ErrNoRoute", ip, info, err)
}
}
// Routes of exactly the minimum length are used.
announce("2001::/16", 64501)
announce("192.0.0.0/8", 64501)
lookup("2001:db8::1", "2001::/16", 64501)
lookup("192.0.2.1", "192.0.0.0/8", 64501)
announce("2001:db8::/32", 64502)
lookup("2001:db8::1", "2001:db8::/32", 64502)
}
// writeOriginRoutes writes one live route from testPeerIP for each prefix in
// origins, originated by the AS it maps to.
func writeOriginRoutes(t *testing.T, db *Database, origins map[string]int) {
t.Helper()
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
routes := make([]*LiveRoute, 0, len(origins))
for prefix, asn := range origins {
routes = append(routes, &LiveRoute{
ID: uuid.New(),
Prefix: prefix,
MaskLength: netip.MustParsePrefix(prefix).Bits(),
IPVersion: detectIPVersion(prefix),
OriginASN: asn,
PeerIP: testPeerIP,
ASPath: []int{asn},
NextHop: testPeerIP,
LastUpdated: ts,
})
}
err := db.UpsertLiveRouteBatch(routes)
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
}
// TestConnectionPoolPragmas holds several pooled connections open at once and
// checks each one carries the per-connection settings from the DSN, plus the
// process-wide hard heap limit, and sees the new file with auto_vacuum
// incremental.
func TestConnectionPoolPragmas(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx := context.Background()
// Hold distinct connections open simultaneously so the pool must open a new
// one (each parsing the DSN) rather than hand back the same connection.
conns := make([]*sql.Conn, 0, heldConnections)
defer func() {
for _, c := range conns {
_ = c.Close()
}
}()
for i := range heldConnections {
c, err := db.db.Conn(ctx)
if err != nil {
t.Fatalf("failed to open connection %d: %v", i, err)
}
conns = append(conns, c)
}
for i, c := range conns {
cacheSize := readPragma(ctx, t, c, i, "cache_size")
if cacheSize != sqliteCacheSizeKiB {
t.Errorf("conn %d: cache_size = %d, want %d",
i, cacheSize, sqliteCacheSizeKiB)
}
busyTimeout := readPragma(ctx, t, c, i, "busy_timeout")
if busyTimeout != sqliteBusyTimeoutMs {
t.Errorf("conn %d: busy_timeout = %d, want %d",
i, busyTimeout, sqliteBusyTimeoutMs)
}
tempStore := readPragma(ctx, t, c, i, "temp_store")
if tempStore == tempStoreMemory {
t.Errorf("conn %d: temp_store = %d, want anything but %d (MEMORY)",
i, tempStore, tempStoreMemory)
}
hardHeapLimit := readPragma(ctx, t, c, i, "hard_heap_limit")
if hardHeapLimit != sqliteHardHeapLimitBytes {
t.Errorf("conn %d: hard_heap_limit = %d, want %d",
i, hardHeapLimit, sqliteHardHeapLimitBytes)
}
autoVacuum := readPragma(ctx, t, c, i, "auto_vacuum")
if autoVacuum != autoVacuumIncremental {
t.Errorf("conn %d: auto_vacuum = %d, want %d (incremental)",
i, autoVacuum, autoVacuumIncremental)
}
}
}
// readPragma reads the integer value of PRAGMA name on connection i, c.
func readPragma(
ctx context.Context, t *testing.T, c *sql.Conn, i int, name string,
) int64 {
t.Helper()
var value int64
err := c.QueryRowContext(ctx, "PRAGMA "+name).Scan(&value)
if err != nil {
t.Fatalf("conn %d: failed to read %s: %v", i, name, err)
}
return value
}
// TestVacuumReturnsFreePages checks that after routes are deleted from a new
// database, one Vacuum call returns every page they used. The deletes leave
// fewer free pages than the 1000 Vacuum frees per call, so none may remain.
// With auto_vacuum off (issue https://git.eeqj.de/sneak/routewatch/issues/43)
// the free pages stayed in the file, and with the PRAGMA run by ExecContext
// Vacuum freed only one page per call.
func TestVacuumReturnsFreePages(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx := context.Background()
ts := time.Now().UTC()
const asn = 64500
routes := make([]*LiveRoute, 0, vacuumTestRoutes)
deletions := make([]LiveRouteDeletion, 0, vacuumTestRoutes)
for i := range vacuumTestRoutes {
route := mkV4Route(t, fmt.Sprintf("10.%d.%d.0/24", i/256, i%256), asn, ts)
routes = append(routes, route)
deletions = append(deletions, LiveRouteDeletion{
Prefix: route.Prefix,
OriginASN: asn,
PeerIP: route.PeerIP,
IPVersion: ipVersionV4,
})
}
err = db.UpsertLiveRouteBatch(routes)
if err != nil {
t.Fatalf("failed to write routes: %v", err)
}
err = db.DeleteLiveRouteBatch(deletions)
if err != nil {
t.Fatalf("failed to delete routes: %v", err)
}
var before int
err = db.db.QueryRowContext(ctx, "PRAGMA freelist_count").Scan(&before)
if err != nil {
t.Fatalf("failed to read freelist_count: %v", err)
}
if before == 0 {
t.Fatalf("no free pages after deleting %d routes", vacuumTestRoutes)
}
err = db.Vacuum(ctx)
if err != nil {
t.Fatalf("Vacuum failed: %v", err)
}
var after int
err = db.db.QueryRowContext(ctx, "PRAGMA freelist_count").Scan(&after)
if err != nil {
t.Fatalf("failed to read freelist_count: %v", err)
}
if after != 0 {
t.Errorf("free pages after Vacuum = %d of %d, want 0", after, before)
}
}
// TestBatchWriteDuringCheckpoint reproduces issue #25. A batch write reads
// (SELECT) before it writes (INSERT/UPDATE). Under the default deferred locking
// the transaction begins as a reader and, when the maintainer's WAL checkpoint
// holds the write lock, its upgrade to writer fails immediately with "database
// is locked" without honouring busy_timeout, dropping the batch. With
// _txlock=immediate the transaction takes the write lock at BEGIN and waits, so
// no batch is dropped. The checkpoint runs without the Database mutex, exactly
// as the background maintainer does in production.
func TestBatchWriteDuringCheckpoint(t *testing.T) {
t.Parallel()
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx, cancel := context.WithCancel(context.Background())
var wg sync.WaitGroup
wg.Add(1)
go func() {
defer wg.Done()
for {
select {
case <-ctx.Done():
return
default:
_ = db.Checkpoint(ctx) // errors are the checkpoint's own to absorb
}
}
}()
ts := time.Now().UTC()
for i := range contentionIterations {
asns := map[int]time.Time{
i % contendedASNCount: ts,
(i % contendedASNCount) + asnSecondBand: ts,
}
err := db.GetOrCreateASNBatch(asns)
if err != nil {
cancel()
wg.Wait()
t.Fatalf("batch write failed under checkpoint contention: %v", err)
}
}
cancel()
wg.Wait()
}
internal/database/database_test.go
-444
View File
@@ -1,444 +0,0 @@
package database
import (
"context"
"database/sql"
"net"
"sync"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/logger"
)
// tempStoreMemory is the PRAGMA temp_store value meaning "hold temp B-trees in
// memory"; the DSN change must leave temp_store below this so they spill to disk.
const tempStoreMemory = 2
// heldConnections is how many pooled connections the pragma test holds open at
// once so each is a distinct SQLite connection that parsed the DSN.
const heldConnections = 5
// Parameters for the checkpoint-contention regression test.
const (
// contentionIterations is how many batch writes race the checkpoint loop.
contentionIterations = 400
// contendedASNCount is the small set of ASNs the batches reuse, so most
// batches update existing rows and exercise the read-before-write path.
contendedASNCount = 16
// asnSecondBand offsets a second ASN per batch so each batch writes more
// than one row.
asnSecondBand = 100
)
func TestIPToUint32(t *testing.T) {
tests := []struct {
name string
ip string
expected uint32
}{
{
name: "Simple IP",
ip: "192.168.1.1",
expected: 3232235777, // 192<<24 + 168<<16 + 1<<8 + 1
},
{
name: "Minimum IP",
ip: "0.0.0.0",
expected: 0,
},
{
name: "Maximum IP",
ip: "255.255.255.255",
expected: 4294967295,
},
{
name: "10.0.0.0",
ip: "10.0.0.0",
expected: 167772160,
},
{
name: "172.16.0.0",
ip: "172.16.0.0",
expected: 2886729728,
},
{
name: "8.8.8.8",
ip: "8.8.8.8",
expected: 134744072,
},
{
name: "1.2.3.4",
ip: "1.2.3.4",
expected: 16909060,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
ip := net.ParseIP(tt.ip)
if ip == nil {
t.Fatalf("Failed to parse IP: %s", tt.ip)
}
result := ipToUint32(ip)
if result != tt.expected {
t.Errorf("ipToUint32(%s) = %d, want %d", tt.ip, result, tt.expected)
}
// Test with IPv4-mapped IPv6 address
ip6 := net.ParseIP(tt.ip).To16()
if ip6 != nil {
result6 := ipToUint32(ip6)
if result6 != tt.expected {
t.Errorf("ipToUint32(%s as IPv6) = %d, want %d", tt.ip, result6, tt.expected)
}
}
})
}
}
func TestCalculateIPv4Range(t *testing.T) {
tests := []struct {
name string
cidr string
wantStart uint32
wantEnd uint32
wantErr bool
}{
{
name: "Single IP /32",
cidr: "192.168.1.1/32",
wantStart: 3232235777,
wantEnd: 3232235777,
},
{
name: "Class C /24",
cidr: "192.168.1.0/24",
wantStart: 3232235776, // 192.168.1.0
wantEnd: 3232236031, // 192.168.1.255
},
{
name: "Class B /16",
cidr: "192.168.0.0/16",
wantStart: 3232235520, // 192.168.0.0
wantEnd: 3232301055, // 192.168.255.255
},
{
name: "Class A /8",
cidr: "10.0.0.0/8",
wantStart: 167772160, // 10.0.0.0
wantEnd: 184549375, // 10.255.255.255
},
{
name: "Entire IPv4 space /0",
cidr: "0.0.0.0/0",
wantStart: 0,
wantEnd: 4294967295,
},
{
name: "Small subnet /30",
cidr: "192.168.1.0/30",
wantStart: 3232235776, // 192.168.1.0
wantEnd: 3232235779, // 192.168.1.3
},
{
name: "Medium subnet /20",
cidr: "172.16.0.0/20",
wantStart: 2886729728, // 172.16.0.0
wantEnd: 2886733823, // 172.16.15.255
},
{
name: "Private range 172.16/12",
cidr: "172.16.0.0/12",
wantStart: 2886729728, // 172.16.0.0
wantEnd: 2887778303, // 172.31.255.255
},
{
name: "Google DNS /29",
cidr: "8.8.8.8/29",
wantStart: 134744072, // 8.8.8.8 (network is actually 8.8.8.8 with /29)
wantEnd: 134744079, // 8.8.8.15
},
{
name: "Non-zero host bits",
cidr: "192.168.1.5/24",
wantStart: 3232235776, // 192.168.1.0 (network address)
wantEnd: 3232236031, // 192.168.1.255
},
{
name: "Invalid CIDR",
cidr: "192.168.1.1/33",
wantErr: true,
},
{
name: "Invalid IP",
cidr: "256.256.256.256/24",
wantErr: true,
},
{
name: "IPv6 CIDR",
cidr: "2001:db8::/32",
wantErr: true,
},
{
name: "Empty CIDR",
cidr: "",
wantErr: true,
},
{
name: "Missing mask",
cidr: "192.168.1.1",
wantErr: true,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
start, end, err := CalculateIPv4Range(tt.cidr)
if tt.wantErr {
if err == nil {
t.Errorf("CalculateIPv4Range(%s) expected error, got nil", tt.cidr)
}
return
}
if err != nil {
t.Errorf("CalculateIPv4Range(%s) unexpected error: %v", tt.cidr, err)
return
}
if start != tt.wantStart {
t.Errorf("CalculateIPv4Range(%s) start = %d, want %d", tt.cidr, start, tt.wantStart)
}
if end != tt.wantEnd {
t.Errorf("CalculateIPv4Range(%s) end = %d, want %d", tt.cidr, end, tt.wantEnd)
}
// Verify that start <= end
if start > end {
t.Errorf("CalculateIPv4Range(%s) start (%d) > end (%d)", tt.cidr, start, end)
}
// Verify the range size matches the CIDR mask
if !tt.wantErr && tt.cidr != "" {
_, ipNet, _ := net.ParseCIDR(tt.cidr)
if ipNet != nil {
ones, bits := ipNet.Mask.Size()
expectedSize := uint32(1) << uint(bits-ones)
actualSize := end - start + 1
if actualSize != expectedSize {
t.Errorf("CalculateIPv4Range(%s) range size = %d, want %d", tt.cidr, actualSize, expectedSize)
}
}
}
})
}
}
func TestIPv4RangeIntegration(t *testing.T) {
// Test that our functions work correctly together
tests := []struct {
name string
cidr string
testIPs []string
shouldContain []bool
}{
{
name: "192.168.1.0/24",
cidr: "192.168.1.0/24",
testIPs: []string{
"192.168.1.0",
"192.168.1.1",
"192.168.1.255",
"192.168.0.255",
"192.168.2.0",
},
shouldContain: []bool{true, true, true, false, false},
},
{
name: "10.0.0.0/8",
cidr: "10.0.0.0/8",
testIPs: []string{
"10.0.0.0",
"10.255.255.255",
"10.1.2.3",
"9.255.255.255",
"11.0.0.0",
},
shouldContain: []bool{true, true, true, false, false},
},
{
name: "172.16.0.0/12",
cidr: "172.16.0.0/12",
testIPs: []string{
"172.16.0.0",
"172.31.255.255",
"172.20.1.1",
"172.15.255.255",
"172.32.0.0",
},
shouldContain: []bool{true, true, true, false, false},
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
start, end, err := CalculateIPv4Range(tt.cidr)
if err != nil {
t.Fatalf("Failed to calculate range for %s: %v", tt.cidr, err)
}
for i, testIP := range tt.testIPs {
ip := net.ParseIP(testIP)
if ip == nil {
t.Fatalf("Failed to parse test IP: %s", testIP)
}
ipUint := ipToUint32(ip)
contained := ipUint >= start && ipUint <= end
if contained != tt.shouldContain[i] {
t.Errorf("IP %s in range %s: got %v, want %v", testIP, tt.cidr, contained, tt.shouldContain[i])
}
}
})
}
}
// TestConnectionPoolPragmas holds several pooled connections open at once and
// checks each one carries the per-connection settings from the DSN, plus the
// process-wide hard heap limit.
func TestConnectionPoolPragmas(t *testing.T) {
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx := context.Background()
// Hold distinct connections open simultaneously so the pool must open a new
// one (each parsing the DSN) rather than hand back the same connection.
conns := make([]*sql.Conn, 0, heldConnections)
defer func() {
for _, c := range conns {
_ = c.Close()
}
}()
for i := 0; i < heldConnections; i++ {
c, err := db.db.Conn(ctx)
if err != nil {
t.Fatalf("failed to open connection %d: %v", i, err)
}
conns = append(conns, c)
}
for i, c := range conns {
var cacheSize int
if err := c.QueryRowContext(ctx, "PRAGMA cache_size").Scan(&cacheSize); err != nil {
t.Fatalf("conn %d: failed to read cache_size: %v", i, err)
}
if cacheSize != sqliteCacheSizeKiB {
t.Errorf("conn %d: cache_size = %d, want %d", i, cacheSize, sqliteCacheSizeKiB)
}
var busyTimeout int
if err := c.QueryRowContext(ctx, "PRAGMA busy_timeout").Scan(&busyTimeout); err != nil {
t.Fatalf("conn %d: failed to read busy_timeout: %v", i, err)
}
if busyTimeout != sqliteBusyTimeoutMs {
t.Errorf("conn %d: busy_timeout = %d, want %d", i, busyTimeout, sqliteBusyTimeoutMs)
}
var tempStore int
if err := c.QueryRowContext(ctx, "PRAGMA temp_store").Scan(&tempStore); err != nil {
t.Fatalf("conn %d: failed to read temp_store: %v", i, err)
}
if tempStore == tempStoreMemory {
t.Errorf("conn %d: temp_store = %d, want anything but %d (MEMORY)", i, tempStore, tempStoreMemory)
}
var hardHeapLimit int64
if err := c.QueryRowContext(ctx, "PRAGMA hard_heap_limit").Scan(&hardHeapLimit); err != nil {
t.Fatalf("conn %d: failed to read hard_heap_limit: %v", i, err)
}
if hardHeapLimit != sqliteHardHeapLimitBytes {
t.Errorf("conn %d: hard_heap_limit = %d, want %d", i, hardHeapLimit, sqliteHardHeapLimitBytes)
}
}
}
// TestBatchWriteDuringCheckpoint reproduces issue #25. A batch write reads
// (SELECT) before it writes (INSERT/UPDATE). Under the default deferred locking
// the transaction begins as a reader and, when the maintainer's WAL checkpoint
// holds the write lock, its upgrade to writer fails immediately with "database
// is locked" without honouring busy_timeout, dropping the batch. With
// _txlock=immediate the transaction takes the write lock at BEGIN and waits, so
// no batch is dropped. The checkpoint runs without the Database mutex, exactly
// as the background maintainer does in production.
func TestBatchWriteDuringCheckpoint(t *testing.T) {
cfg := &config.Config{StateDir: t.TempDir()}
db, err := New(cfg, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
ctx, cancel := context.WithCancel(context.Background())
var wg sync.WaitGroup
wg.Add(1)
go func() {
defer wg.Done()
for {
select {
case <-ctx.Done():
return
default:
_ = db.Checkpoint(ctx) // errors are the checkpoint's own to absorb
}
}
}()
ts := time.Now().UTC()
for i := 0; i < contentionIterations; i++ {
asns := map[int]time.Time{
i % contendedASNCount: ts,
(i % contendedASNCount) + asnSecondBand: ts,
}
if err := db.GetOrCreateASNBatch(asns); err != nil {
cancel()
wg.Wait()
t.Fatalf("batch write failed under checkpoint contention: %v", err)
}
}
cancel()
wg.Wait()
}
func BenchmarkIPToUint32(b *testing.B) {
ip := net.ParseIP("192.168.1.1")
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = ipToUint32(ip)
}
}
func BenchmarkCalculateIPv4Range(b *testing.B) {
cidr := "192.168.0.0/16"
b.ResetTimer()
for i := 0; i < b.N; i++ {
_, _, _ = CalculateIPv4Range(cidr)
}
}
internal/database/interface.go
+15 -7
View File
@@ -18,6 +18,8 @@ type Stats struct {
Peers int Peers int
FileSizeBytes int64 FileSizeBytes int64
LiveRoutes int LiveRoutes int
IPv4Routes int
IPv6Routes int
OldestRoute *time.Time OldestRoute *time.Time
NewestRoute *time.Time NewestRoute *time.Time
IPv4PrefixDistribution []PrefixDistribution IPv4PrefixDistribution []PrefixDistribution
@@ -27,6 +29,8 @@ type Stats struct {
// Store defines the interface for database operations. It provides methods for // Store defines the interface for database operations. It provides methods for
// managing ASNs, prefixes, announcements, peerings, BGP peers, and live routes. // managing ASNs, prefixes, announcements, peerings, BGP peers, and live routes.
// Implementations must be safe for concurrent use. // Implementations must be safe for concurrent use.
//
//nolint:interfacebloat // the one seam handlers and tests depend on
type Store interface { type Store interface {
// ASN operations // ASN operations
GetOrCreateASN(number int, timestamp time.Time) (*ASN, error) GetOrCreateASN(number int, timestamp time.Time) (*ASN, error)
@@ -55,14 +59,14 @@ type Store interface {
UpsertLiveRouteBatch(routes []*LiveRoute) error UpsertLiveRouteBatch(routes []*LiveRoute) error
DeleteLiveRoute(prefix string, originASN int, peerIP string) error DeleteLiveRoute(prefix string, originASN int, peerIP string) error
DeleteLiveRouteBatch(deletions []LiveRouteDeletion) error DeleteLiveRouteBatch(deletions []LiveRouteDeletion) error
GetPrefixDistribution() (ipv4 []PrefixDistribution, ipv6 []PrefixDistribution, err error) GetPrefixDistribution() (
GetPrefixDistributionContext(ctx context.Context) (ipv4 []PrefixDistribution, ipv6 []PrefixDistribution, err error) ipv4 []PrefixDistribution, ipv6 []PrefixDistribution, err error)
GetPrefixDistributionContext(ctx context.Context) (
ipv4 []PrefixDistribution, ipv6 []PrefixDistribution, err error)
GetLiveRouteCounts() (ipv4Count, ipv6Count int, err error) GetLiveRouteCounts() (ipv4Count, ipv6Count int, err error)
GetLiveRouteCountsContext(ctx context.Context) (ipv4Count, ipv6Count int, err error) GetLiveRouteCountsContext(ctx context.Context) (ipv4Count, ipv6Count int, err error)
// IP lookup operations // IP lookup operations
GetASInfoForIP(ip string) (*ASInfo, error)
GetASInfoForIPContext(ctx context.Context, ip string) (*ASInfo, error)
GetIPInfo(ip string) (*IPInfo, error) GetIPInfo(ip string) (*IPInfo, error)
GetIPInfoContext(ctx context.Context, ip string) (*IPInfo, error) GetIPInfoContext(ctx context.Context, ip string) (*IPInfo, error)
@@ -76,10 +80,14 @@ type Store interface {
GetASDetailsContext(ctx context.Context, asn int) (*ASN, []LiveRoute, error) GetASDetailsContext(ctx context.Context, asn int) (*ASN, []LiveRoute, error)
GetASPeers(asn int) ([]ASPeer, error) GetASPeers(asn int) ([]ASPeer, error)
GetASPeersContext(ctx context.Context, asn int) ([]ASPeer, error) GetASPeersContext(ctx context.Context, asn int) ([]ASPeer, error)
GetPrefixDetails(prefix string) ([]LiveRoute, error) GetPrefixDetails(prefix string) ([]LiveRoute, map[int]ASN, error)
GetPrefixDetailsContext(ctx context.Context, prefix string) ([]LiveRoute, error) GetPrefixDetailsContext(
ctx context.Context, prefix string,
) ([]LiveRoute, map[int]ASN, error)
GetRandomPrefixesByLength(maskLength, ipVersion, limit int) ([]LiveRoute, error) GetRandomPrefixesByLength(maskLength, ipVersion, limit int) ([]LiveRoute, error)
GetRandomPrefixesByLengthContext(ctx context.Context, maskLength, ipVersion, limit int) ([]LiveRoute, error) GetRandomPrefixesByLengthContext(
ctx context.Context, maskLength, ipVersion, limit int,
) ([]LiveRoute, error)
// Lifecycle // Lifecycle
Close() error Close() error
internal/database/models.go
+26 -32
View File
@@ -1,4 +1,3 @@
// Package database provides SQLite storage for BGP routing data.
package database package database
import ( import (
@@ -9,6 +8,8 @@ import (
// ASN represents an Autonomous System Number with its metadata including // ASN represents an Autonomous System Number with its metadata including
// handle, description, WHOIS data, and first/last seen timestamps. // handle, description, WHOIS data, and first/last seen timestamps.
//
//nolint:tagliatelle // field names are the existing JSON API
type ASN struct { type ASN struct {
ASN int `json:"asn"` ASN int `json:"asn"`
Handle string `json:"handle"` Handle string `json:"handle"`
@@ -36,37 +37,39 @@ type ASN struct {
// Prefix represents an IP prefix (CIDR block) with its IP version (4 or 6) // Prefix represents an IP prefix (CIDR block) with its IP version (4 or 6)
// and first/last seen timestamps. // and first/last seen timestamps.
type Prefix struct { type Prefix struct {
ID uuid.UUID `json:"id"` ID uuid.UUID
Prefix string `json:"prefix"` Prefix string
IPVersion int `json:"ip_version"` // 4 or 6 IPVersion int // 4 or 6
FirstSeen time.Time `json:"first_seen"` FirstSeen time.Time
LastSeen time.Time `json:"last_seen"` LastSeen time.Time
} }
// Announcement represents a BGP announcement or withdrawal event, // Announcement represents a BGP announcement or withdrawal event,
// containing the prefix, AS path, origin ASN, peer ASN, next hop, and timestamp. // containing the prefix, AS path, origin ASN, peer ASN, next hop, and timestamp.
type Announcement struct { type Announcement struct {
ID uuid.UUID `json:"id"` ID uuid.UUID
PrefixID uuid.UUID `json:"prefix_id"` PrefixID uuid.UUID
PeerASN int `json:"peer_asn"` PeerASN int
OriginASN int `json:"origin_asn"` OriginASN int
Path string `json:"path"` // JSON-encoded AS path Path string // JSON-encoded AS path
NextHop string `json:"next_hop"` NextHop string
Timestamp time.Time `json:"timestamp"` Timestamp time.Time
IsWithdrawal bool `json:"is_withdrawal"` IsWithdrawal bool
} }
// ASNPeering represents a peering relationship between two ASNs, // ASNPeering represents a peering relationship between two ASNs,
// stored with the lower ASN as ASA and the higher as ASB. // stored with the lower ASN as ASA and the higher as ASB.
type ASNPeering struct { type ASNPeering struct {
ID uuid.UUID `json:"id"` ID uuid.UUID
ASA int `json:"as_a"` ASA int
ASB int `json:"as_b"` ASB int
FirstSeen time.Time `json:"first_seen"` FirstSeen time.Time
LastSeen time.Time `json:"last_seen"` LastSeen time.Time
} }
// LiveRoute represents a route in the live routing table // LiveRoute represents a route in the live routing table
//
//nolint:tagliatelle // field names are the existing JSON API
type LiveRoute struct { type LiveRoute struct {
ID uuid.UUID `json:"id"` ID uuid.UUID `json:"id"`
Prefix string `json:"prefix"` Prefix string `json:"prefix"`
@@ -77,28 +80,19 @@ type LiveRoute struct {
ASPath []int `json:"as_path"` ASPath []int `json:"as_path"`
NextHop string `json:"next_hop"` NextHop string `json:"next_hop"`
LastUpdated time.Time `json:"last_updated"` LastUpdated time.Time `json:"last_updated"`
// IPv4 range fields for fast lookups (nil for IPv6)
V4IPStart *uint32 `json:"v4_ip_start,omitempty"`
V4IPEnd *uint32 `json:"v4_ip_end,omitempty"`
} }
// PrefixDistribution represents the distribution of prefixes by mask length // PrefixDistribution represents the distribution of prefixes by mask length
//
//nolint:tagliatelle // field names are the existing JSON API
type PrefixDistribution struct { type PrefixDistribution struct {
MaskLength int `json:"mask_length"` MaskLength int `json:"mask_length"`
Count int `json:"count"` Count int `json:"count"`
} }
// ASInfo represents AS information for an IP lookup (legacy format)
type ASInfo struct {
ASN int `json:"asn"`
Handle string `json:"handle"`
Description string `json:"description"`
Prefix string `json:"prefix"`
LastUpdated time.Time `json:"last_updated"`
Age string `json:"age"`
}
// IPInfo represents comprehensive IP information for the /ip endpoint // IPInfo represents comprehensive IP information for the /ip endpoint
//
//nolint:tagliatelle // field names are the existing JSON API
type IPInfo struct { type IPInfo struct {
IP string `json:"ip"` IP string `json:"ip"`
PTR []string `json:"ptr,omitempty"` PTR []string `json:"ptr,omitempty"`
internal/database/schema.sql
-6
View File
@@ -107,9 +107,6 @@ CREATE TABLE IF NOT EXISTS live_routes_v4 (
as_path TEXT NOT NULL, -- JSON array as_path TEXT NOT NULL, -- JSON array
next_hop TEXT NOT NULL, next_hop TEXT NOT NULL,
last_updated DATETIME NOT NULL, last_updated DATETIME NOT NULL,
-- IPv4 range columns for fast lookups
ip_start INTEGER NOT NULL, -- Start of IPv4 range as 32-bit unsigned int
ip_end INTEGER NOT NULL, -- End of IPv4 range as 32-bit unsigned int
UNIQUE(prefix, origin_asn, peer_ip) UNIQUE(prefix, origin_asn, peer_ip)
); );
@@ -123,7 +120,6 @@ CREATE TABLE IF NOT EXISTS live_routes_v6 (
as_path TEXT NOT NULL, -- JSON array as_path TEXT NOT NULL, -- JSON array
next_hop TEXT NOT NULL, next_hop TEXT NOT NULL,
last_updated DATETIME NOT NULL, last_updated DATETIME NOT NULL,
-- Note: IPv6 doesn't use integer range columns
UNIQUE(prefix, origin_asn, peer_ip) UNIQUE(prefix, origin_asn, peer_ip)
); );
@@ -132,8 +128,6 @@ CREATE INDEX IF NOT EXISTS idx_live_routes_v4_prefix ON live_routes_v4(prefix);
CREATE INDEX IF NOT EXISTS idx_live_routes_v4_mask_length ON live_routes_v4(mask_length); CREATE INDEX IF NOT EXISTS idx_live_routes_v4_mask_length ON live_routes_v4(mask_length);
CREATE INDEX IF NOT EXISTS idx_live_routes_v4_origin_asn ON live_routes_v4(origin_asn); CREATE INDEX IF NOT EXISTS idx_live_routes_v4_origin_asn ON live_routes_v4(origin_asn);
CREATE INDEX IF NOT EXISTS idx_live_routes_v4_last_updated ON live_routes_v4(last_updated); CREATE INDEX IF NOT EXISTS idx_live_routes_v4_last_updated ON live_routes_v4(last_updated);
-- Indexes for IPv4 range queries
CREATE INDEX IF NOT EXISTS idx_live_routes_v4_ip_range ON live_routes_v4(ip_start, ip_end);
-- Index to optimize prefix distribution queries -- Index to optimize prefix distribution queries
CREATE INDEX IF NOT EXISTS idx_live_routes_v4_mask_prefix ON live_routes_v4(mask_length, prefix); CREATE INDEX IF NOT EXISTS idx_live_routes_v4_mask_prefix ON live_routes_v4(mask_length, prefix);
internal/database/slowquery.go
+17 -36
View File
@@ -19,29 +19,21 @@ func logSlowQuery(logger *logger.Logger, query string, start time.Time) {
} }
// queryRow wraps QueryRow with slow query logging // queryRow wraps QueryRow with slow query logging
// nolint:unused // kept for consistency with other query wrappers //
func (d *Database) queryRow(query string, args ...interface{}) *sql.Row { //nolint:unused // kept for consistency with other query wrappers
func (d *Database) queryRow(query string, args ...any) *sql.Row {
start := time.Now() start := time.Now()
defer logSlowQuery(d.logger, query, start) defer logSlowQuery(d.logger, query, start)
return d.db.QueryRow(query, args...) return d.db.QueryRowContext(context.Background(), query, args...)
}
// query wraps Query with slow query logging
// nolint:unused // kept for future use to ensure all queries go through slow query logging
func (d *Database) query(query string, args ...interface{}) (*sql.Rows, error) {
start := time.Now()
defer logSlowQuery(d.logger, query, start)
return d.db.Query(query, args...)
} }
// exec wraps Exec with slow query logging // exec wraps Exec with slow query logging
func (d *Database) exec(query string, args ...interface{}) error { func (d *Database) exec(query string, args ...any) error {
start := time.Now() start := time.Now()
defer logSlowQuery(d.logger, query, start) defer logSlowQuery(d.logger, query, start)
_, err := d.db.Exec(query, args...) _, err := d.db.ExecContext(context.Background(), query, args...)
return err return err
} }
@@ -49,35 +41,22 @@ func (d *Database) exec(query string, args ...interface{}) error {
// loggingTx wraps sql.Tx to log slow queries // loggingTx wraps sql.Tx to log slow queries
type loggingTx struct { type loggingTx struct {
*sql.Tx *sql.Tx
logger *logger.Logger logger *logger.Logger
} }
// QueryRow wraps sql.Tx.QueryRow to log slow queries // QueryRow wraps sql.Tx.QueryRow to log slow queries
func (tx *loggingTx) QueryRow(query string, args ...interface{}) *sql.Row { func (tx *loggingTx) QueryRow(query string, args ...any) *sql.Row {
start := time.Now() start := time.Now()
defer logSlowQuery(tx.logger, query, start) defer logSlowQuery(tx.logger, query, start)
return tx.Tx.QueryRow(query, args...) return tx.Tx.QueryRowContext(context.Background(), query, args...)
}
// Query wraps sql.Tx.Query to log slow queries
func (tx *loggingTx) Query(query string, args ...interface{}) (*sql.Rows, error) {
start := time.Now()
defer logSlowQuery(tx.logger, query, start)
return tx.Tx.Query(query, args...)
}
// QueryContext wraps sql.Tx.QueryContext to log slow queries
func (tx *loggingTx) QueryContext(ctx context.Context, query string, args ...interface{}) (*sql.Rows, error) {
start := time.Now()
defer logSlowQuery(tx.logger, query, start)
return tx.Tx.QueryContext(ctx, query, args...)
} }
// QueryRowContext wraps sql.Tx.QueryRowContext to log slow queries // QueryRowContext wraps sql.Tx.QueryRowContext to log slow queries
func (tx *loggingTx) QueryRowContext(ctx context.Context, query string, args ...interface{}) *sql.Row { func (tx *loggingTx) QueryRowContext(
ctx context.Context, query string, args ...any,
) *sql.Row {
start := time.Now() start := time.Now()
defer logSlowQuery(tx.logger, query, start) defer logSlowQuery(tx.logger, query, start)
@@ -85,15 +64,17 @@ func (tx *loggingTx) QueryRowContext(ctx context.Context, query string, args ...
} }
// Exec wraps sql.Tx.Exec to log slow queries // Exec wraps sql.Tx.Exec to log slow queries
func (tx *loggingTx) Exec(query string, args ...interface{}) (sql.Result, error) { func (tx *loggingTx) Exec(query string, args ...any) (sql.Result, error) {
start := time.Now() start := time.Now()
defer logSlowQuery(tx.logger, query, start) defer logSlowQuery(tx.logger, query, start)
return tx.Tx.Exec(query, args...) return tx.Tx.ExecContext(context.Background(), query, args...)
} }
// ExecContext wraps sql.Tx.ExecContext to log slow queries // ExecContext wraps sql.Tx.ExecContext to log slow queries
func (tx *loggingTx) ExecContext(ctx context.Context, query string, args ...interface{}) (sql.Result, error) { func (tx *loggingTx) ExecContext(
ctx context.Context, query string, args ...any,
) (sql.Result, error) {
start := time.Now() start := time.Now()
defer logSlowQuery(tx.logger, query, start) defer logSlowQuery(tx.logger, query, start)
internal/database/utils.go
+14
View File
@@ -1,6 +1,8 @@
package database package database
import ( import (
"fmt"
"net"
"strings" "strings"
"github.com/google/uuid" "github.com/google/uuid"
@@ -18,3 +20,15 @@ func detectIPVersion(prefix string) int {
return ipVersionV4 return ipVersionV4
} }
// prefixMaskLength returns the mask length of a prefix such as 192.0.2.0/24.
func prefixMaskLength(prefix string) (int, error) {
_, network, err := net.ParseCIDR(prefix)
if err != nil {
return 0, fmt.Errorf("invalid prefix %s: %w", prefix, err)
}
maskLength, _ := network.Mask.Size()
return maskLength, nil
}
internal/logger/logger.go
+1
View File
@@ -74,6 +74,7 @@ func getSourceAttrs() []slog.Attr {
// Get the function name // Get the function name
fn := runtime.FuncForPC(pc) fn := runtime.FuncForPC(pc)
var funcName string var funcName string
if fn != nil { if fn != nil {
funcName = filepath.Base(fn.Name()) funcName = filepath.Base(fn.Name())
internal/metrics/metrics.go
+25 -14
View File
@@ -60,17 +60,23 @@ func New() *Tracker {
} }
} }
// SetConnected updates the connection status // SetConnected updates the connection status. Every connection after the
// first counts one reconnect.
func (t *Tracker) SetConnected(connected bool) { func (t *Tracker) SetConnected(connected bool) {
wasConnected := t.isConnected.Swap(connected) t.isConnected.Store(connected)
if connected {
t.mu.Lock() if !connected {
t.connectedSince = time.Now() return
t.mu.Unlock() }
// Increment reconnect count (but not for the initial connection)
if wasConnected || t.reconnectCount.Load() > 0 { t.mu.Lock()
t.reconnectCount.Add(1) // connectedSince is zero until the first connection.
} connectedBefore := !t.connectedSince.IsZero()
t.connectedSince = time.Now()
t.mu.Unlock()
if connectedBefore {
t.reconnectCount.Add(1)
} }
} }
@@ -115,9 +121,11 @@ func (t *Tracker) GetStreamMetrics() StreamMetrics {
if msgCount >= 0 { if msgCount >= 0 {
totalMessages = uint64(msgCount) totalMessages = uint64(msgCount)
} }
if byteCount >= 0 { if byteCount >= 0 {
totalBytes = uint64(byteCount) totalBytes = uint64(byteCount)
} }
if wireByteCount >= 0 { if wireByteCount >= 0 {
totalWireBytes = uint64(wireByteCount) totalWireBytes = uint64(wireByteCount)
} }
@@ -161,7 +169,7 @@ func (t *Tracker) RecordWithdrawal() {
func (t *Tracker) SetBGPPeerCount(count int) { func (t *Tracker) SetBGPPeerCount(count int) {
// BGP peer count is always small (< 1000), so int32 is safe // BGP peer count is always small (< 1000), so int32 is safe
if count > 0 && count < 1<<31 { if count > 0 && count < 1<<31 {
t.bgpPeerCount.Store(int32(count)) //nolint:gosec // count is validated t.bgpPeerCount.Store(int32(count))
} }
} }
@@ -209,7 +217,8 @@ type StreamMetrics struct {
TotalMessages uint64 TotalMessages uint64
// TotalBytes is the total number of decompressed bytes received since startup // TotalBytes is the total number of decompressed bytes received since startup
TotalBytes uint64 TotalBytes uint64
// TotalWireBytes is the total number of bytes received on the wire (before decompression) // TotalWireBytes is the total number of bytes received on the wire (before
// decompression)
TotalWireBytes uint64 TotalWireBytes uint64
// ConnectedSince is the time when the current connection was established // ConnectedSince is the time when the current connection was established
ConnectedSince time.Time ConnectedSince time.Time
@@ -217,9 +226,11 @@ type StreamMetrics struct {
Connected bool Connected bool
// MessagesPerSec is the rate of messages received per second (1-minute average) // MessagesPerSec is the rate of messages received per second (1-minute average)
MessagesPerSec float64 MessagesPerSec float64
// BitsPerSec is the rate of decompressed bits received per second (1-minute average) // BitsPerSec is the rate of decompressed bits received per second (1-minute
// average)
BitsPerSec float64 BitsPerSec float64
// WireBitsPerSec is the rate of bits received on the wire per second (1-minute average) // WireBitsPerSec is the rate of bits received on the wire per second
// (1-minute average)
WireBitsPerSec float64 WireBitsPerSec float64
// ReconnectCount is the number of reconnections since startup // ReconnectCount is the number of reconnections since startup
ReconnectCount uint64 ReconnectCount uint64
internal/metrics/metrics_test.go
+27
View File
@@ -0,0 +1,27 @@
package metrics_test
import (
"testing"
"git.eeqj.de/sneak/routewatch/internal/metrics"
)
// TestReconnectCount connects three times with a disconnect between each, the
// way the streamer does, and expects every connection after the first to count
// one reconnect.
func TestReconnectCount(t *testing.T) {
t.Parallel()
tracker := metrics.New()
tracker.SetConnected(true)
tracker.SetConnected(false)
tracker.SetConnected(true)
tracker.SetConnected(false)
tracker.SetConnected(true)
const wantReconnects = 2
if got := tracker.GetReconnectCount(); got != wantReconnects {
t.Fatalf("reconnect count = %d, want %d", got, wantReconnects)
}
}
internal/ristypes/ris.go
+15 -5
View File
@@ -1,4 +1,5 @@
// Package ristypes defines the data structures for RIS Live BGP messages and announcements. // Package ristypes defines the data structures for RIS Live BGP messages and
// announcements.
package ristypes package ristypes
import ( import (
@@ -17,25 +18,30 @@ type ASPath []int
func (p *ASPath) UnmarshalJSON(data []byte) error { func (p *ASPath) UnmarshalJSON(data []byte) error {
// First try to unmarshal as a simple array of integers // First try to unmarshal as a simple array of integers
var simple []int var simple []int
if err := json.Unmarshal(data, &simple); err == nil {
err := json.Unmarshal(data, &simple)
if err == nil {
*p = ASPath(simple) *p = ASPath(simple)
return nil return nil
} }
// If that fails, unmarshal as array of interfaces and flatten // If that fails, unmarshal as array of interfaces and flatten
var raw []interface{} var raw []any
if err := json.Unmarshal(data, &raw); err != nil {
err = json.Unmarshal(data, &raw)
if err != nil {
return err return err
} }
// Flatten the array // Flatten the array
result := make([]int, 0) result := make([]int, 0)
for _, item := range raw { for _, item := range raw {
switch v := item.(type) { switch v := item.(type) {
case float64: case float64:
result = append(result, int(v)) result = append(result, int(v))
case []interface{}: case []any:
// Nested array - flatten it // Nested array - flatten it
for _, nested := range v { for _, nested := range v {
if num, ok := nested.(float64); ok { if num, ok := nested.(float64); ok {
@@ -62,6 +68,8 @@ type RISLiveMessage struct {
// It contains metadata about the BGP session (peer, ASN, host) along with // It contains metadata about the BGP session (peer, ASN, host) along with
// the actual BGP update data including AS path, communities, announcements, // the actual BGP update data including AS path, communities, announcements,
// and withdrawals. // and withdrawals.
//
//nolint:tagliatelle // field names are the RIS Live feed's
type RISMessage struct { type RISMessage struct {
Type string `json:"type"` Type string `json:"type"`
Timestamp float64 `json:"timestamp"` Timestamp float64 `json:"timestamp"`
@@ -90,6 +98,8 @@ type RISMessage struct {
// RISAnnouncement represents a BGP route announcement within a RIS message. // RISAnnouncement represents a BGP route announcement within a RIS message.
// It contains the next hop IP address and the list of prefixes being announced // It contains the next hop IP address and the list of prefixes being announced
// via that next hop. // via that next hop.
//
//nolint:tagliatelle // field names are the RIS Live feed's
type RISAnnouncement struct { type RISAnnouncement struct {
NextHop string `json:"next_hop"` NextHop string `json:"next_hop"`
Prefixes []string `json:"prefixes"` Prefixes []string `json:"prefixes"`
internal/ristypes/ris_test.go
+24 -7
View File
@@ -1,10 +1,12 @@
package ristypes package ristypes_test
import ( import (
"bufio" "bufio"
"encoding/json" "encoding/json"
"os" "os"
"testing" "testing"
"git.eeqj.de/sneak/routewatch/internal/ristypes"
) )
// messageExamplesPath is the captured RIS Live feed used as decode fixtures, // messageExamplesPath is the captured RIS Live feed used as decode fixtures,
@@ -15,32 +17,43 @@ const messageExamplesPath = "../../docs/message-examples.json"
// streamer does and checks that the fields no handler reads (Community, Raw) // streamer does and checks that the fields no handler reads (Community, Raw)
// stay empty while the fields handlers use (Path, Announcements) still decode. // stay empty while the fields handlers use (Path, Announcements) still decode.
func TestDecodeDropsCommunityAndRaw(t *testing.T) { func TestDecodeDropsCommunityAndRaw(t *testing.T) {
t.Parallel()
f, err := os.Open(messageExamplesPath) f, err := os.Open(messageExamplesPath)
if err != nil { if err != nil {
t.Fatalf("open fixtures: %v", err) t.Fatalf("open fixtures: %v", err)
} }
defer f.Close()
defer func() {
_ = f.Close()
}()
var messages, withPath, withAnnouncements int var messages, withPath, withAnnouncements int
scanner := bufio.NewScanner(f) scanner := bufio.NewScanner(f)
scanner.Buffer(make([]byte, 0, 64*1024), 1024*1024) scanner.Buffer(make([]byte, 0, 64*1024), 1024*1024)
for scanner.Scan() { for scanner.Scan() {
line := scanner.Bytes() line := scanner.Bytes()
if len(line) == 0 { if len(line) == 0 {
continue continue
} }
var wrapper RISLiveMessage var wrapper ristypes.RISLiveMessage
if err := json.Unmarshal(line, &wrapper); err != nil {
err := json.Unmarshal(line, &wrapper)
if err != nil {
t.Fatalf("unmarshal message %d: %v", messages+1, err) t.Fatalf("unmarshal message %d: %v", messages+1, err)
} }
messages++ messages++
msg := wrapper.Data msg := wrapper.Data
if msg.Community != nil { if msg.Community != nil {
t.Errorf("message %d: Community decoded, want empty: %v", messages, msg.Community) t.Errorf("message %d: Community decoded, want empty: %v",
messages, msg.Community)
} }
if msg.Raw != "" { if msg.Raw != "" {
t.Errorf("message %d: Raw decoded, want empty", messages) t.Errorf("message %d: Raw decoded, want empty", messages)
} }
@@ -48,11 +61,14 @@ func TestDecodeDropsCommunityAndRaw(t *testing.T) {
if len(msg.Path) > 0 { if len(msg.Path) > 0 {
withPath++ withPath++
} }
if len(msg.Announcements) > 0 { if len(msg.Announcements) > 0 {
withAnnouncements++ withAnnouncements++
} }
} }
if err := scanner.Err(); err != nil {
err = scanner.Err()
if err != nil {
t.Fatalf("scan fixtures: %v", err) t.Fatalf("scan fixtures: %v", err)
} }
@@ -62,6 +78,7 @@ func TestDecodeDropsCommunityAndRaw(t *testing.T) {
// The fixtures include announcement messages; a used field must still decode, // The fixtures include announcement messages; a used field must still decode,
// otherwise an empty Community/Raw would prove nothing. // otherwise an empty Community/Raw would prove nothing.
if withPath == 0 || withAnnouncements == 0 { if withPath == 0 || withAnnouncements == 0 {
t.Fatalf("used fields did not decode: withPath=%d withAnnouncements=%d", withPath, withAnnouncements) t.Fatalf("used fields did not decode: withPath=%d withAnnouncements=%d",
withPath, withAnnouncements)
} }
} }
internal/routewatch/app.go
+32 -16
View File
@@ -1,10 +1,11 @@
// Package routewatch contains the primary RouteWatch type that represents a running instance // Package routewatch contains the primary RouteWatch type that represents a
// of the application and contains pointers to its core dependencies, and is responsible for initialization. // running instance of the application and contains pointers to its core
// dependencies, and is responsible for initialization.
package routewatch package routewatch
import ( import (
"context" "context"
"fmt" "errors"
"sync" "sync"
"time" "time"
@@ -18,6 +19,10 @@ import (
"go.uber.org/fx" "go.uber.org/fx"
) )
// errNonBatchedNotImplemented is returned by Run when batched database writes
// are disabled, since only the batched handlers exist.
var errNonBatchedNotImplemented = errors.New("non-batched handlers not implemented")
// Dependencies contains all dependencies for RouteWatch // Dependencies contains all dependencies for RouteWatch
type Dependencies struct { type Dependencies struct {
fx.In fx.In
@@ -68,8 +73,10 @@ func (rw *RouteWatch) Run(ctx context.Context) error {
// Apply runtime limit if specified // Apply runtime limit if specified
if rw.maxRuntime > 0 { if rw.maxRuntime > 0 {
var cancel context.CancelFunc var cancel context.CancelFunc
ctx, cancel = context.WithTimeout(ctx, rw.maxRuntime) ctx, cancel = context.WithTimeout(ctx, rw.maxRuntime)
defer cancel() defer cancel()
rw.logger.Info("Running with time limit", "max_runtime", rw.maxRuntime) rw.logger.Info("Running with time limit", "max_runtime", rw.maxRuntime)
} }
@@ -96,29 +103,31 @@ func (rw *RouteWatch) Run(ctx context.Context) error {
// Non-batched handlers not implemented yet // Non-batched handlers not implemented yet
rw.logger.Error("Non-batched handlers not implemented") rw.logger.Error("Non-batched handlers not implemented")
return fmt.Errorf("non-batched handlers not implemented") return errNonBatchedNotImplemented
} }
// No longer need routing table handler - PrefixHandler maintains live_routes table // No longer need routing table handler - PrefixHandler maintains live_routes table
// Start streaming // Start streaming
if err := rw.streamer.Start(); err != nil { err := rw.streamer.Start(ctx)
if err != nil {
return err return err
} }
// Start HTTP server // Start HTTP server
if err := rw.server.Start(); err != nil { err = rw.server.Start(ctx)
if err != nil {
return err return err
} }
// Start ASN WHOIS fetcher for background updates // Start ASN WHOIS fetcher for background updates
rw.asnFetcher = NewASNFetcher(rw.db, rw.logger.Logger) rw.asnFetcher = NewASNFetcher(rw.db, rw.logger.Logger)
rw.asnFetcher.Start() rw.asnFetcher.Start(ctx)
rw.server.SetASNFetcher(rw.asnFetcher) rw.server.SetASNFetcher(rw.asnFetcher)
// Start database maintenance goroutine // Start database maintenance goroutine
rw.dbMaintainer = NewDBMaintainer(rw.db, rw.logger.Logger) rw.dbMaintainer = NewDBMaintainer(rw.db, rw.logger.Logger)
rw.dbMaintainer.Start() rw.dbMaintainer.Start(ctx)
// Wait for context cancellation // Wait for context cancellation
<-ctx.Done() <-ctx.Done()
@@ -127,13 +136,14 @@ func (rw *RouteWatch) Run(ctx context.Context) error {
} }
// Shutdown performs graceful shutdown of all services // Shutdown performs graceful shutdown of all services
func (rw *RouteWatch) Shutdown() { func (rw *RouteWatch) Shutdown(ctx context.Context) {
rw.mu.Lock() rw.mu.Lock()
if rw.shutdown { if rw.shutdown {
rw.mu.Unlock() rw.mu.Unlock()
return return
} }
rw.shutdown = true rw.shutdown = true
rw.mu.Unlock() rw.mu.Unlock()
@@ -142,14 +152,17 @@ func (rw *RouteWatch) Shutdown() {
rw.logger.Info("Flushing database handler") rw.logger.Info("Flushing database handler")
rw.dbHandler.Stop() rw.dbHandler.Stop()
} }
if rw.peerHandler != nil { if rw.peerHandler != nil {
rw.logger.Info("Flushing peer handler") rw.logger.Info("Flushing peer handler")
rw.peerHandler.Stop() rw.peerHandler.Stop()
} }
if rw.prefixHandler != nil { if rw.prefixHandler != nil {
rw.logger.Info("Flushing prefix handler") rw.logger.Info("Flushing prefix handler")
rw.prefixHandler.Stop() rw.prefixHandler.Stop()
} }
if rw.peeringHandler != nil { if rw.peeringHandler != nil {
rw.logger.Info("Flushing peering handler") rw.logger.Info("Flushing peering handler")
rw.peeringHandler.Stop() rw.peeringHandler.Stop()
@@ -168,11 +181,15 @@ func (rw *RouteWatch) Shutdown() {
// Stop services // Stop services
rw.streamer.Stop() rw.streamer.Stop()
// Stop HTTP server with a timeout // Stop HTTP server with a timeout of its own, however much of ctx is left
const serverStopTimeout = 5 * time.Second const serverStopTimeout = 5 * time.Second
stopCtx, cancel := context.WithTimeout(context.Background(), serverStopTimeout)
stopCtx, cancel := context.WithTimeout(
context.WithoutCancel(ctx), serverStopTimeout)
defer cancel() defer cancel()
if err := rw.server.Stop(stopCtx); err != nil {
err := rw.server.Stop(stopCtx)
if err != nil {
rw.logger.Error("Failed to stop HTTP server gracefully", "error", err) rw.logger.Error("Failed to stop HTTP server gracefully", "error", err)
} }
@@ -185,12 +202,11 @@ func (rw *RouteWatch) Shutdown() {
"bits_per_sec", metrics.BitsPerSec, "bits_per_sec", metrics.BitsPerSec,
"duration", time.Since(metrics.ConnectedSince), "duration", time.Since(metrics.ConnectedSince),
) )
} }
// getModule provides all fx dependencies // getModule provides all fx dependencies
func getModule() fx.Option { func getModule() []fx.Option {
return fx.Options( return []fx.Option{
fx.Provide( fx.Provide(
logger.New, logger.New,
config.New, config.New,
@@ -203,5 +219,5 @@ func getModule() fx.Option {
server.New, server.New,
New, New,
), ),
) }
} }
internal/routewatch/app_integration_test.go → internal/routewatch/app_integration_internal_test.go
+77 -52
View File
@@ -50,7 +50,9 @@ func newMockStore() *mockStore {
} }
// GetOrCreateASN mock implementation // GetOrCreateASN mock implementation
func (m *mockStore) GetOrCreateASN(number int, timestamp time.Time) (*database.ASN, error) { func (m *mockStore) GetOrCreateASN(
number int, timestamp time.Time,
) (*database.ASN, error) {
m.mu.Lock() m.mu.Lock()
defer m.mu.Unlock() defer m.mu.Unlock()
@@ -99,11 +101,14 @@ func (m *mockStore) UpdatePrefixesBatch(prefixes map[string]time.Time) error {
} }
} }
} }
return nil return nil
} }
// GetOrCreatePrefix mock implementation // GetOrCreatePrefix mock implementation
func (m *mockStore) GetOrCreatePrefix(prefix string, timestamp time.Time) (*database.Prefix, error) { func (m *mockStore) GetOrCreatePrefix(
prefix string, timestamp time.Time,
) (*database.Prefix, error) {
m.mu.Lock() m.mu.Lock()
defer m.mu.Unlock() defer m.mu.Unlock()
@@ -168,7 +173,7 @@ func (m *mockStore) RecordPeering(asA, asB int, _ time.Time) error {
} }
// UpdatePeer mock implementation // UpdatePeer mock implementation
func (m *mockStore) UpdatePeer(peerIP string, peerASN int, messageType string, timestamp time.Time) error { func (m *mockStore) UpdatePeer(_ string, _ int, _ string, _ time.Time) error {
// Simple mock - just return nil // Simple mock - just return nil
return nil return nil
} }
@@ -194,63 +199,48 @@ func (m *mockStore) GetStats() (database.Stats, error) {
} }
// GetStatsContext returns statistics about the mock store with context support // GetStatsContext returns statistics about the mock store with context support
func (m *mockStore) GetStatsContext(ctx context.Context) (database.Stats, error) { func (m *mockStore) GetStatsContext(_ context.Context) (database.Stats, error) {
return m.GetStats() return m.GetStats()
} }
// UpsertLiveRoute mock implementation // UpsertLiveRoute mock implementation
func (m *mockStore) UpsertLiveRoute(route *database.LiveRoute) error { func (m *mockStore) UpsertLiveRoute(_ *database.LiveRoute) error {
// Simple mock - just return nil // Simple mock - just return nil
return nil return nil
} }
// DeleteLiveRoute mock implementation // DeleteLiveRoute mock implementation
func (m *mockStore) DeleteLiveRoute(prefix string, originASN int, peerIP string) error { func (m *mockStore) DeleteLiveRoute(_ string, _ int, _ string) error {
// Simple mock - just return nil // Simple mock - just return nil
return nil return nil
} }
// GetPrefixDistribution mock implementation // GetPrefixDistribution mock implementation
func (m *mockStore) GetPrefixDistribution() (ipv4 []database.PrefixDistribution, ipv6 []database.PrefixDistribution, err error) { func (m *mockStore) GetPrefixDistribution() (
[]database.PrefixDistribution, []database.PrefixDistribution, error,
) {
// Return empty distributions for now // Return empty distributions for now
return nil, nil, nil return nil, nil, nil
} }
// GetPrefixDistributionContext mock implementation with context support // GetPrefixDistributionContext mock implementation with context support
func (m *mockStore) GetPrefixDistributionContext(ctx context.Context) (ipv4 []database.PrefixDistribution, ipv6 []database.PrefixDistribution, err error) { func (m *mockStore) GetPrefixDistributionContext(_ context.Context) (
[]database.PrefixDistribution, []database.PrefixDistribution, error,
) {
return m.GetPrefixDistribution() return m.GetPrefixDistribution()
} }
// GetLiveRouteCounts mock implementation // GetLiveRouteCounts mock implementation
func (m *mockStore) GetLiveRouteCounts() (ipv4Count, ipv6Count int, err error) { func (m *mockStore) GetLiveRouteCounts() (int, int, error) {
// Return mock counts // Return mock counts
return m.RouteCount / 2, m.RouteCount / 2, nil return m.RouteCount / 2, m.RouteCount / 2, nil
} }
// GetLiveRouteCountsContext mock implementation with context support // GetLiveRouteCountsContext mock implementation with context support
func (m *mockStore) GetLiveRouteCountsContext(ctx context.Context) (ipv4Count, ipv6Count int, err error) { func (m *mockStore) GetLiveRouteCountsContext(_ context.Context) (int, int, error) {
return m.GetLiveRouteCounts() return m.GetLiveRouteCounts()
} }
// GetASInfoForIP mock implementation
func (m *mockStore) GetASInfoForIP(ip string) (*database.ASInfo, error) {
// Simple mock - return a test AS
now := time.Now()
return &database.ASInfo{
ASN: 15169,
Handle: "GOOGLE",
Description: "Google LLC",
Prefix: "8.8.8.0/24",
LastUpdated: now.Add(-5 * time.Minute),
Age: "5m0s",
}, nil
}
// GetASInfoForIPContext mock implementation with context support
func (m *mockStore) GetASInfoForIPContext(ctx context.Context, ip string) (*database.ASInfo, error) {
return m.GetASInfoForIP(ip)
}
// GetASDetails mock implementation // GetASDetails mock implementation
func (m *mockStore) GetASDetails(asn int) (*database.ASN, []database.LiveRoute, error) { func (m *mockStore) GetASDetails(asn int) (*database.ASN, []database.LiveRoute, error) {
m.mu.Lock() m.mu.Lock()
@@ -266,39 +256,51 @@ func (m *mockStore) GetASDetails(asn int) (*database.ASN, []database.LiveRoute,
} }
// GetASDetailsContext mock implementation with context support // GetASDetailsContext mock implementation with context support
func (m *mockStore) GetASDetailsContext(ctx context.Context, asn int) (*database.ASN, []database.LiveRoute, error) { func (m *mockStore) GetASDetailsContext(
_ context.Context, asn int,
) (*database.ASN, []database.LiveRoute, error) {
return m.GetASDetails(asn) return m.GetASDetails(asn)
} }
// GetPrefixDetails mock implementation // GetPrefixDetails mock implementation
func (m *mockStore) GetPrefixDetails(prefix string) ([]database.LiveRoute, error) { func (m *mockStore) GetPrefixDetails(
_ string,
) ([]database.LiveRoute, map[int]database.ASN, error) {
// Return empty routes for now // Return empty routes for now
return []database.LiveRoute{}, nil return []database.LiveRoute{}, map[int]database.ASN{}, nil
} }
// GetPrefixDetailsContext mock implementation with context support // GetPrefixDetailsContext mock implementation with context support
func (m *mockStore) GetPrefixDetailsContext(ctx context.Context, prefix string) ([]database.LiveRoute, error) { func (m *mockStore) GetPrefixDetailsContext(
_ context.Context, prefix string,
) ([]database.LiveRoute, map[int]database.ASN, error) {
return m.GetPrefixDetails(prefix) return m.GetPrefixDetails(prefix)
} }
func (m *mockStore) GetRandomPrefixesByLength(maskLength, ipVersion, limit int) ([]database.LiveRoute, error) { func (m *mockStore) GetRandomPrefixesByLength(
_, _, _ int,
) ([]database.LiveRoute, error) {
// Return empty routes for now // Return empty routes for now
return []database.LiveRoute{}, nil return []database.LiveRoute{}, nil
} }
// GetRandomPrefixesByLengthContext mock implementation with context support // GetRandomPrefixesByLengthContext mock implementation with context support
func (m *mockStore) GetRandomPrefixesByLengthContext(ctx context.Context, maskLength, ipVersion, limit int) ([]database.LiveRoute, error) { func (m *mockStore) GetRandomPrefixesByLengthContext(
_ context.Context, maskLength, ipVersion, limit int,
) ([]database.LiveRoute, error) {
return m.GetRandomPrefixesByLength(maskLength, ipVersion, limit) return m.GetRandomPrefixesByLength(maskLength, ipVersion, limit)
} }
// GetASPeers mock implementation // GetASPeers mock implementation
func (m *mockStore) GetASPeers(asn int) ([]database.ASPeer, error) { func (m *mockStore) GetASPeers(_ int) ([]database.ASPeer, error) {
// Return empty peers for now // Return empty peers for now
return []database.ASPeer{}, nil return []database.ASPeer{}, nil
} }
// GetASPeersContext mock implementation with context support // GetASPeersContext mock implementation with context support
func (m *mockStore) GetASPeersContext(ctx context.Context, asn int) ([]database.ASPeer, error) { func (m *mockStore) GetASPeersContext(
_ context.Context, asn int,
) ([]database.ASPeer, error) {
return m.GetASPeers(asn) return m.GetASPeers(asn)
} }
@@ -308,8 +310,11 @@ func (m *mockStore) GetIPInfo(ip string) (*database.IPInfo, error) {
} }
// GetIPInfoContext mock implementation with context support // GetIPInfoContext mock implementation with context support
func (m *mockStore) GetIPInfoContext(ctx context.Context, ip string) (*database.IPInfo, error) { func (m *mockStore) GetIPInfoContext(
_ context.Context, ip string,
) (*database.IPInfo, error) {
now := time.Now() now := time.Now()
return &database.IPInfo{ return &database.IPInfo{
IP: ip, IP: ip,
Netblock: "8.8.8.0/24", Netblock: "8.8.8.0/24",
@@ -326,17 +331,21 @@ func (m *mockStore) GetIPInfoContext(ctx context.Context, ip string) (*database.
} }
// GetNextStaleASN mock implementation // GetNextStaleASN mock implementation
func (m *mockStore) GetNextStaleASN(ctx context.Context, staleThreshold time.Duration) (int, error) { func (m *mockStore) GetNextStaleASN(_ context.Context, _ time.Duration) (int, error) {
return 0, database.ErrNoStaleASN return 0, database.ErrNoStaleASN
} }
// UpdateASNWHOIS mock implementation // UpdateASNWHOIS mock implementation
func (m *mockStore) UpdateASNWHOIS(ctx context.Context, update *database.ASNWHOISUpdate) error { func (m *mockStore) UpdateASNWHOIS(
_ context.Context, _ *database.ASNWHOISUpdate,
) error {
return nil return nil
} }
// GetWHOISStats mock implementation // GetWHOISStats mock implementation
func (m *mockStore) GetWHOISStats(ctx context.Context, staleThreshold time.Duration) (*database.WHOISStats, error) { func (m *mockStore) GetWHOISStats(
_ context.Context, _ time.Duration,
) (*database.WHOISStats, error) {
m.mu.Lock() m.mu.Lock()
defer m.mu.Unlock() defer m.mu.Unlock()
@@ -364,19 +373,22 @@ func (m *mockStore) UpsertLiveRouteBatch(routes []*database.LiveRoute) error {
LastSeen: route.LastUpdated, LastSeen: route.LastUpdated,
} }
m.PrefixCount++ m.PrefixCount++
if route.IPVersion == 4 { if route.IPVersion == 4 {
m.IPv4Prefixes++ m.IPv4Prefixes++
} else { } else {
m.IPv6Prefixes++ m.IPv6Prefixes++
} }
} }
m.RouteCount++ m.RouteCount++
} }
return nil return nil
} }
// DeleteLiveRouteBatch mock implementation // DeleteLiveRouteBatch mock implementation
func (m *mockStore) DeleteLiveRouteBatch(deletions []database.LiveRouteDeletion) error { func (m *mockStore) DeleteLiveRouteBatch(_ []database.LiveRouteDeletion) error {
// Simple mock - just return nil // Simple mock - just return nil
return nil return nil
} }
@@ -396,43 +408,53 @@ func (m *mockStore) GetOrCreateASNBatch(asns map[int]time.Time) error {
m.ASNCount++ m.ASNCount++
} }
} }
return nil return nil
} }
// UpdatePeerBatch mock implementation // UpdatePeerBatch mock implementation
func (m *mockStore) UpdatePeerBatch(peers map[string]database.PeerUpdate) error { func (m *mockStore) UpdatePeerBatch(_ map[string]database.PeerUpdate) error {
// Simple mock - just return nil // Simple mock - just return nil
return nil return nil
} }
// Vacuum mock implementation // Vacuum mock implementation
func (m *mockStore) Vacuum(ctx context.Context) error { func (m *mockStore) Vacuum(_ context.Context) error {
return nil return nil
} }
// Analyze mock implementation // Analyze mock implementation
func (m *mockStore) Analyze(ctx context.Context) error { func (m *mockStore) Analyze(_ context.Context) error {
return nil return nil
} }
// Checkpoint mock implementation // Checkpoint mock implementation
func (m *mockStore) Checkpoint(ctx context.Context) error { func (m *mockStore) Checkpoint(_ context.Context) error {
return nil return nil
} }
// Ping mock implementation // Ping mock implementation
func (m *mockStore) Ping(ctx context.Context) error { func (m *mockStore) Ping(_ context.Context) error {
return nil return nil
} }
func TestRouteWatchLiveFeed(t *testing.T) { func TestRouteWatchLiveFeed(t *testing.T) {
t.Parallel()
if testing.Short() { if testing.Short() {
t.Skip("skipping live RIPE RIS network feed test in short mode; run without -short to include it") t.Skip("skipping live RIPE RIS network feed test in short mode; " +
"run without -short to include it")
} }
// Create mock database // Create mock database
mockDB := newMockStore() mockDB := newMockStore()
defer mockDB.Close()
defer func() {
err := mockDB.Close()
if err != nil {
t.Errorf("failed to close mock database: %v", err)
}
}()
logger := logger.New() logger := logger.New()
@@ -450,7 +472,7 @@ func TestRouteWatchLiveFeed(t *testing.T) {
} }
// Create server // Create server
srv := server.New(mockDB, s, logger) srv := server.New(mockDB, s, logger, cfg)
// Create RouteWatch with 5 second limit // Create RouteWatch with 5 second limit
deps := Dependencies{ deps := Dependencies{
@@ -485,16 +507,19 @@ func TestRouteWatchLiveFeed(t *testing.T) {
if stats.ASNs == 0 { if stats.ASNs == 0 {
t.Error("Expected to receive some ASNs from live feed") t.Error("Expected to receive some ASNs from live feed")
} }
t.Logf("Received %d unique ASNs in 5 seconds", stats.ASNs) t.Logf("Received %d unique ASNs in 5 seconds", stats.ASNs)
if stats.Prefixes == 0 { if stats.Prefixes == 0 {
t.Error("Expected to receive some prefixes from live feed") t.Error("Expected to receive some prefixes from live feed")
} }
t.Logf("Received %d unique prefixes (%d IPv4, %d IPv6) in 5 seconds", stats.Prefixes, stats.IPv4Prefixes, stats.IPv6Prefixes)
t.Logf("Received %d unique prefixes (%d IPv4, %d IPv6) in 5 seconds",
stats.Prefixes, stats.IPv4Prefixes, stats.IPv6Prefixes)
if stats.Peerings == 0 { if stats.Peerings == 0 {
t.Error("Expected to receive some peerings from live feed") t.Error("Expected to receive some peerings from live feed")
} }
t.Logf("Recorded %d AS peering relationships in 5 seconds", stats.Peerings)
t.Logf("Recorded %d AS peering relationships in 5 seconds", stats.Peerings)
} }
internal/routewatch/app_test.go
+1 -1
View File
@@ -1,3 +1,3 @@
package routewatch package routewatch_test
// Tests for routewatch package are in app_integration_test.go // Tests for routewatch package are in app_integration_test.go
internal/routewatch/ashandler.go
+18 -14
View File
@@ -71,7 +71,7 @@ func NewASHandler(db database.Store, logger *logger.Logger) *ASHandler {
// the AS path information needed to track autonomous systems. // the AS path information needed to track autonomous systems.
func (h *ASHandler) WantsMessage(messageType string) bool { func (h *ASHandler) WantsMessage(messageType string) bool {
// We only care about UPDATE messages for the database // We only care about UPDATE messages for the database
return messageType == "UPDATE" return messageType == updateMessageType
} }
// QueueCapacity returns the recommended message queue size for this handler. // QueueCapacity returns the recommended message queue size for this handler.
@@ -124,9 +124,20 @@ func (h *ASHandler) HandleMessage(msg *ristypes.RISMessage) {
} }
} }
// Stop gracefully shuts down the ASHandler by signaling the background flush
// goroutine to terminate and waiting for it to complete. Any pending ASN
// operations in the current batch are flushed to the database before Stop
// returns. This method should be called during application shutdown to ensure
// no data is lost.
func (h *ASHandler) Stop() {
close(h.stopCh)
h.wg.Wait()
}
// flushLoop runs in a goroutine and periodically flushes batches // flushLoop runs in a goroutine and periodically flushes batches
func (h *ASHandler) flushLoop() { func (h *ASHandler) flushLoop() {
defer h.wg.Done() defer h.wg.Done()
ticker := time.NewTicker(asnBatchTimeout) ticker := time.NewTicker(asnBatchTimeout)
defer ticker.Stop() defer ticker.Stop()
@@ -149,7 +160,8 @@ func (h *ASHandler) flushLoop() {
} }
} }
// flushBatchLocked flushes the ASN batch to the database (must be called with mutex held) // flushBatchLocked flushes the ASN batch to the database (must be called with
// mutex held)
func (h *ASHandler) flushBatchLocked() { func (h *ASHandler) flushBatchLocked() {
if len(h.batch) == 0 { if len(h.batch) == 0 {
return return
@@ -164,21 +176,13 @@ func (h *ASHandler) flushBatchLocked() {
} }
// Process all ASNs in a single batch transaction // Process all ASNs in a single batch transaction
if err := h.db.GetOrCreateASNBatch(asnMap); err != nil { err := h.db.GetOrCreateASNBatch(asnMap)
h.logger.Error("Failed to process ASN batch", "error", err, "count", len(asnMap)) if err != nil {
h.logger.Error("Failed to process ASN batch",
"error", err, "count", len(asnMap))
} }
// Clear batch // Clear batch
h.batch = h.batch[:0] h.batch = h.batch[:0]
h.lastFlush = time.Now() h.lastFlush = time.Now()
} }
// Stop gracefully shuts down the ASHandler by signaling the background flush
// goroutine to terminate and waiting for it to complete. Any pending ASN
// operations in the current batch are flushed to the database before Stop
// returns. This method should be called during application shutdown to ensure
// no data is lost.
func (h *ASHandler) Stop() {
close(h.stopCh)
h.wg.Wait()
}
internal/routewatch/asnfetcher.go
+21 -24
View File
@@ -1,8 +1,8 @@
// Package routewatch contains the ASN WHOIS fetcher for background updates.
package routewatch package routewatch
import ( import (
"context" "context"
"errors"
"log/slog" "log/slog"
"sync" "sync"
"time" "time"
@@ -73,10 +73,13 @@ func NewASNFetcher(db database.Store, logger *slog.Logger) *ASNFetcher {
} }
} }
// Start begins the background ASN fetcher goroutine. // Start begins the background ASN fetcher goroutine. It runs until Stop:
func (f *ASNFetcher) Start() { // the end of ctx does not stop it.
func (f *ASNFetcher) Start(ctx context.Context) {
f.wg.Add(1) f.wg.Add(1)
go f.run()
go f.run(context.WithoutCancel(ctx))
f.logger.Info("ASN fetcher started", f.logger.Info("ASN fetcher started",
"base_interval", baseInterval, "base_interval", baseInterval,
"min_interval", minInterval, "min_interval", minInterval,
@@ -151,10 +154,7 @@ func (f *ASNFetcher) recordSuccess() {
f.consecutiveFails = 0 f.consecutiveFails = 0
// Decrease interval by half, but not below minimum // Decrease interval by half, but not below minimum
newInterval := f.currentInterval / backoffMultiplier newInterval := max(f.currentInterval/backoffMultiplier, minInterval)
if newInterval < minInterval {
newInterval = minInterval
}
if newInterval != f.currentInterval { if newInterval != f.currentInterval {
f.logger.Debug("Decreased fetch interval", f.logger.Debug("Decreased fetch interval",
@@ -177,10 +177,7 @@ func (f *ASNFetcher) recordFailure() {
f.consecutiveFails++ f.consecutiveFails++
// Exponential backoff: multiply by 2, capped at max // Exponential backoff: multiply by 2, capped at max
newInterval := f.currentInterval * backoffMultiplier newInterval := min(f.currentInterval*backoffMultiplier, maxInterval)
if newInterval > maxInterval {
newInterval = maxInterval
}
if newInterval != f.currentInterval { if newInterval != f.currentInterval {
f.logger.Debug("Increased fetch interval due to failure", f.logger.Debug("Increased fetch interval due to failure",
@@ -199,7 +196,7 @@ func (f *ASNFetcher) recordFailure() {
} }
// run is the main background loop. // run is the main background loop.
func (f *ASNFetcher) run() { func (f *ASNFetcher) run(ctx context.Context) {
defer f.wg.Done() defer f.wg.Done()
timer := time.NewTimer(f.getInterval()) timer := time.NewTimer(f.getInterval())
@@ -212,13 +209,13 @@ func (f *ASNFetcher) run() {
case asn := <-f.immediateQueue: case asn := <-f.immediateQueue:
// Process immediate request (respects lock) // Process immediate request (respects lock)
f.tryFetch(asn) f.tryFetch(ctx, asn)
// Reset timer after immediate fetch // Reset timer after immediate fetch
timer.Reset(f.getInterval()) timer.Reset(f.getInterval())
case <-timer.C: case <-timer.C:
// Background fetch of stale/missing ASN // Background fetch of stale/missing ASN
f.fetchNextStale() f.fetchNextStale(ctx)
// Reset timer with potentially updated interval // Reset timer with potentially updated interval
timer.Reset(f.getInterval()) timer.Reset(f.getInterval())
} }
@@ -227,7 +224,7 @@ func (f *ASNFetcher) run() {
// tryFetch attempts to fetch and update an ASN, respecting the fetch lock. // tryFetch attempts to fetch and update an ASN, respecting the fetch lock.
// Returns true if fetch was successful. // Returns true if fetch was successful.
func (f *ASNFetcher) tryFetch(asn int) bool { func (f *ASNFetcher) tryFetch(ctx context.Context, asn int) bool {
// Try to acquire lock, skip if another fetch is running // Try to acquire lock, skip if another fetch is running
if !f.fetchMu.TryLock() { if !f.fetchMu.TryLock() {
f.logger.Debug("Skipping fetch, another fetch in progress", "asn", asn) f.logger.Debug("Skipping fetch, another fetch in progress", "asn", asn)
@@ -236,11 +233,11 @@ func (f *ASNFetcher) tryFetch(asn int) bool {
} }
defer f.fetchMu.Unlock() defer f.fetchMu.Unlock()
return f.fetchAndUpdate(asn) return f.fetchAndUpdate(ctx, asn)
} }
// fetchNextStale finds and fetches the next ASN needing WHOIS data. // fetchNextStale finds and fetches the next ASN needing WHOIS data.
func (f *ASNFetcher) fetchNextStale() { func (f *ASNFetcher) fetchNextStale(ctx context.Context) {
// Try to acquire lock, skip if another fetch is running // Try to acquire lock, skip if another fetch is running
if !f.fetchMu.TryLock() { if !f.fetchMu.TryLock() {
f.logger.Debug("Skipping stale fetch, another fetch in progress") f.logger.Debug("Skipping stale fetch, another fetch in progress")
@@ -249,12 +246,12 @@ func (f *ASNFetcher) fetchNextStale() {
} }
defer f.fetchMu.Unlock() defer f.fetchMu.Unlock()
ctx, cancel := context.WithTimeout(context.Background(), time.Minute) lookupCtx, cancel := context.WithTimeout(ctx, time.Minute)
defer cancel() defer cancel()
asn, err := f.db.GetNextStaleASN(ctx, whoisStaleThreshold) asn, err := f.db.GetNextStaleASN(lookupCtx, whoisStaleThreshold)
if err != nil { if err != nil {
if err != database.ErrNoStaleASN { if !errors.Is(err, database.ErrNoStaleASN) {
f.logger.Error("Failed to get stale ASN", "error", err) f.logger.Error("Failed to get stale ASN", "error", err)
f.recordFailure() f.recordFailure()
} }
@@ -263,13 +260,13 @@ func (f *ASNFetcher) fetchNextStale() {
return return
} }
f.fetchAndUpdate(asn) f.fetchAndUpdate(ctx, asn)
} }
// fetchAndUpdate performs a WHOIS lookup and updates the database. // fetchAndUpdate performs a WHOIS lookup and updates the database.
// Returns true if successful. // Returns true if successful.
func (f *ASNFetcher) fetchAndUpdate(asn int) bool { func (f *ASNFetcher) fetchAndUpdate(ctx context.Context, asn int) bool {
ctx, cancel := context.WithTimeout(context.Background(), time.Minute) ctx, cancel := context.WithTimeout(ctx, time.Minute)
defer cancel() defer cancel()
f.logger.Info("Fetching WHOIS data", "asn", asn) f.logger.Info("Fetching WHOIS data", "asn", asn)
internal/routewatch/cli.go
+25 -9
View File
@@ -20,6 +20,8 @@ const (
debugInterval = 60 * time.Second debugInterval = 60 * time.Second
// bytesPerMB is bytes per megabyte // bytesPerMB is bytes per megabyte
bytesPerMB = 1024 * 1024 bytesPerMB = 1024 * 1024
// failureExitCode is the exit status when RouteWatch stops on an error
failureExitCode = 1
) )
// logDebugStats logs goroutine count and memory usage // logDebugStats logs goroutine count and memory usage
@@ -55,14 +57,19 @@ func logDebugStats(logger *logger.Logger) {
// CLIEntry is the main entry point for the routewatch command-line interface. // CLIEntry is the main entry point for the routewatch command-line interface.
// It initializes the application using the fx dependency injection framework, // It initializes the application using the fx dependency injection framework,
// sets up signal handling for graceful shutdown, and starts the RouteWatch service. // sets up signal handling for graceful shutdown, and starts the RouteWatch
// This function blocks until the application receives a shutdown signal or encounters // service. This function blocks until the application receives a shutdown
// a fatal error. // signal or encounters a fatal error.
func CLIEntry() { func CLIEntry() {
app := fx.New( app := fx.New(
getModule(), fx.Options(getModule()...),
fx.StopTimeout(shutdownTimeout), // Allow 60 seconds for graceful shutdown fx.StopTimeout(shutdownTimeout), // Allow 60 seconds for graceful shutdown
fx.Invoke(func(lc fx.Lifecycle, rw *RouteWatch, logger *logger.Logger, shutdowner fx.Shutdowner) { fx.Invoke(func(
lc fx.Lifecycle,
rw *RouteWatch,
logger *logger.Logger,
shutdowner fx.Shutdowner,
) {
lc.Append(fx.Hook{ lc.Append(fx.Hook{
OnStart: func(ctx context.Context) error { OnStart: func(ctx context.Context) error {
// Start debug stats logging // Start debug stats logging
@@ -75,22 +82,31 @@ func CLIEntry() {
go func() { go func() {
<-sigCh <-sigCh
logger.Info("Received shutdown signal") logger.Info("Received shutdown signal")
if err := shutdowner.Shutdown(); err != nil {
err := shutdowner.Shutdown()
if err != nil {
logger.Error("Failed to shutdown gracefully", "error", err) logger.Error("Failed to shutdown gracefully", "error", err)
} }
}() }()
go func() { go func() {
if err := rw.Run(ctx); err != nil { err := rw.Run(ctx)
if err != nil {
logger.Error("RouteWatch error", "error", err) logger.Error("RouteWatch error", "error", err)
err = shutdowner.Shutdown(fx.ExitCode(failureExitCode))
if err != nil {
logger.Error("Failed to shutdown gracefully",
"error", err)
}
} }
}() }()
return nil return nil
}, },
OnStop: func(_ context.Context) error { OnStop: func(ctx context.Context) error {
logger.Info("Shutting down RouteWatch") logger.Info("Shutting down RouteWatch")
rw.Shutdown() rw.Shutdown(ctx)
return nil return nil
}, },
internal/routewatch/dbmaintainer.go
+61 -59
View File
@@ -1,4 +1,3 @@
// Package routewatch contains the database maintainer for background maintenance tasks.
package routewatch package routewatch
import ( import (
@@ -64,10 +63,13 @@ func NewDBMaintainer(db database.Store, logger *slog.Logger) *DBMaintainer {
} }
} }
// Start begins the background maintenance goroutine. // Start begins the background maintenance goroutine. It runs until Stop:
func (m *DBMaintainer) Start() { // the end of ctx does not stop it.
func (m *DBMaintainer) Start(ctx context.Context) {
m.wg.Add(1) m.wg.Add(1)
go m.run()
go m.run(context.WithoutCancel(ctx))
m.logger.Info("Database maintainer started", m.logger.Info("Database maintainer started",
"checkpoint_interval", checkpointInterval, "checkpoint_interval", checkpointInterval,
"vacuum_interval", vacuumInterval, "vacuum_interval", vacuumInterval,
@@ -83,15 +85,17 @@ func (m *DBMaintainer) Stop() {
} }
// run is the main background loop. // run is the main background loop.
func (m *DBMaintainer) run() { func (m *DBMaintainer) run(ctx context.Context) {
defer m.wg.Done() defer m.wg.Done()
// Use different timers for each task // Use different timers for each task
checkpointTimer := time.NewTimer(checkpointInterval) checkpointTimer := time.NewTimer(checkpointInterval)
vacuumTimer := time.NewTimer(vacuumInterval)
analyzeTimer := time.NewTimer(analyzeInterval)
defer checkpointTimer.Stop() defer checkpointTimer.Stop()
vacuumTimer := time.NewTimer(vacuumInterval)
defer vacuumTimer.Stop() defer vacuumTimer.Stop()
analyzeTimer := time.NewTimer(analyzeInterval)
defer analyzeTimer.Stop() defer analyzeTimer.Stop()
for { for {
@@ -100,90 +104,88 @@ func (m *DBMaintainer) run() {
return return
case <-checkpointTimer.C: case <-checkpointTimer.C:
m.runCheckpoint() m.runCheckpoint(ctx)
checkpointTimer.Reset(checkpointInterval) checkpointTimer.Reset(checkpointInterval)
case <-vacuumTimer.C: case <-vacuumTimer.C:
m.runVacuum() m.runVacuum(ctx)
vacuumTimer.Reset(vacuumInterval) vacuumTimer.Reset(vacuumInterval)
case <-analyzeTimer.C: case <-analyzeTimer.C:
m.runAnalyze() m.runAnalyze(ctx)
analyzeTimer.Reset(analyzeInterval) analyzeTimer.Reset(analyzeInterval)
} }
} }
} }
// runCheckpoint performs a WAL checkpoint to keep the WAL file small. // runCheckpoint performs a WAL checkpoint to keep the WAL file small.
func (m *DBMaintainer) runCheckpoint() { func (m *DBMaintainer) runCheckpoint(ctx context.Context) {
ctx, cancel := context.WithTimeout(context.Background(), checkpointTimeout) duration, err := runWithTimeout(ctx, checkpointTimeout, m.db.Checkpoint)
defer cancel() m.recordResult(&m.lastCheckpoint, &m.lastCheckpointError, &m.checkpointCount, err)
startTime := time.Now()
err := m.db.Checkpoint(ctx)
m.statsMu.Lock()
m.lastCheckpoint = time.Now()
m.lastCheckpointError = err
if err == nil {
m.checkpointCount++
}
m.statsMu.Unlock()
if err != nil { if err != nil {
m.logger.Error("WAL checkpoint failed", "error", err, "duration", time.Since(startTime)) m.logger.Error("WAL checkpoint failed", "error", err, "duration", duration)
} else { } else {
m.logger.Debug("WAL checkpoint completed", "duration", time.Since(startTime)) m.logger.Debug("WAL checkpoint completed", "duration", duration)
} }
} }
// runVacuum performs an incremental vacuum operation on the database. // runVacuum performs an incremental vacuum operation on the database.
func (m *DBMaintainer) runVacuum() { func (m *DBMaintainer) runVacuum(ctx context.Context) {
ctx, cancel := context.WithTimeout(context.Background(), vacuumTimeout)
defer cancel()
m.logger.Debug("Running incremental vacuum") m.logger.Debug("Running incremental vacuum")
startTime := time.Now()
err := m.db.Vacuum(ctx) duration, err := runWithTimeout(ctx, vacuumTimeout, m.db.Vacuum)
m.recordResult(&m.lastVacuum, &m.lastVacuumError, &m.vacuumCount, err)
m.statsMu.Lock()
m.lastVacuum = time.Now()
m.lastVacuumError = err
if err == nil {
m.vacuumCount++
}
m.statsMu.Unlock()
if err != nil { if err != nil {
m.logger.Error("Incremental vacuum failed", "error", err, "duration", time.Since(startTime)) m.logger.Error("Incremental vacuum failed", "error", err, "duration", duration)
} else { } else {
m.logger.Debug("Incremental vacuum completed", "duration", time.Since(startTime)) m.logger.Debug("Incremental vacuum completed", "duration", duration)
} }
} }
// runAnalyze performs an ANALYZE operation on the database. // runAnalyze performs an ANALYZE operation on the database.
func (m *DBMaintainer) runAnalyze() { func (m *DBMaintainer) runAnalyze(ctx context.Context) {
ctx, cancel := context.WithTimeout(context.Background(), analyzeTimeout)
defer cancel()
m.logger.Info("Starting database ANALYZE") m.logger.Info("Starting database ANALYZE")
startTime := time.Now()
err := m.db.Analyze(ctx) duration, err := runWithTimeout(ctx, analyzeTimeout, m.db.Analyze)
m.recordResult(&m.lastAnalyze, &m.lastAnalyzeError, &m.analyzeCount, err)
m.statsMu.Lock()
m.lastAnalyze = time.Now()
m.lastAnalyzeError = err
if err == nil {
m.analyzeCount++
}
m.statsMu.Unlock()
if err != nil { if err != nil {
m.logger.Error("ANALYZE failed", "error", err, "duration", time.Since(startTime)) m.logger.Error("ANALYZE failed", "error", err, "duration", duration)
} else { } else {
m.logger.Info("ANALYZE completed", "duration", time.Since(startTime)) m.logger.Info("ANALYZE completed", "duration", duration)
}
}
// runWithTimeout runs operation with a context that ends after timeout and
// returns how long it took and its error.
func runWithTimeout(
ctx context.Context,
timeout time.Duration,
operation func(context.Context) error,
) (time.Duration, error) {
ctx, cancel := context.WithTimeout(ctx, timeout)
defer cancel()
startTime := time.Now()
err := operation(ctx)
return time.Since(startTime), err
}
// recordResult stores when a maintenance task last ran and its error, and
// counts it when it succeeded.
func (m *DBMaintainer) recordResult(
lastRun *time.Time, lastErr *error, successCount *int, err error,
) {
m.statsMu.Lock()
defer m.statsMu.Unlock()
*lastRun = time.Now()
*lastErr = err
if err == nil {
*successCount++
} }
} }
internal/routewatch/handler.go
+16 -14
View File
@@ -1,10 +1,16 @@
package routewatch package routewatch
import ( import (
"slices"
"git.eeqj.de/sneak/routewatch/internal/logger" "git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/ristypes" "git.eeqj.de/sneak/routewatch/internal/ristypes"
) )
// updateMessageType is the RIS message type of BGP UPDATE messages, the only
// type the database handlers process.
const updateMessageType = "UPDATE"
// SimpleHandler is a basic implementation of streamer.MessageHandler that // SimpleHandler is a basic implementation of streamer.MessageHandler that
// filters messages by type and delegates processing to a callback function. // filters messages by type and delegates processing to a callback function.
// It provides a simple way to handle specific RIS message types without // It provides a simple way to handle specific RIS message types without
@@ -15,10 +21,11 @@ type SimpleHandler struct {
callback func(*ristypes.RISMessage) callback func(*ristypes.RISMessage)
} }
// NewSimpleHandler creates a new SimpleHandler that accepts specific message types. // NewSimpleHandler creates a new SimpleHandler that accepts specific message
// The messageTypes parameter specifies which RIS message types this handler will process. // types. The messageTypes parameter specifies which RIS message types this
// If messageTypes is empty, the handler will accept all message types. // handler will process. If messageTypes is empty, the handler will accept all
// The callback function is invoked for each message that passes the type filter. // message types. The callback function is invoked for each message that
// passes the type filter.
func NewSimpleHandler( func NewSimpleHandler(
logger *logger.Logger, logger *logger.Logger,
messageTypes []string, messageTypes []string,
@@ -31,22 +38,17 @@ func NewSimpleHandler(
} }
} }
// WantsMessage returns true if this handler wants to process messages of the given type. // WantsMessage returns true if this handler wants to process messages of the
// It checks whether messageType is in the handler's configured list of accepted types. // given type. It checks whether messageType is in the handler's configured
// If no specific types were configured (empty messageTypes slice), it returns true for all types. // list of accepted types. If no specific types were configured (empty
// messageTypes slice), it returns true for all types.
func (h *SimpleHandler) WantsMessage(messageType string) bool { func (h *SimpleHandler) WantsMessage(messageType string) bool {
// If no specific types are set, accept all messages // If no specific types are set, accept all messages
if len(h.messageTypes) == 0 { if len(h.messageTypes) == 0 {
return true return true
} }
for _, t := range h.messageTypes { return slices.Contains(h.messageTypes, messageType)
if t == messageType {
return true
}
}
return false
} }
// HandleMessage processes a RIS message by invoking the configured callback function. // HandleMessage processes a RIS message by invoking the configured callback function.
internal/routewatch/peerhandler.go
+21 -13
View File
@@ -1,7 +1,8 @@
package routewatch package routewatch
// peerhandler.go provides batched peer tracking functionality for BGP route monitoring. // peerhandler.go provides batched peer tracking functionality for BGP route
// It tracks BGP peers from all incoming RIS messages and maintains peer state in the database. // monitoring. It tracks BGP peers from all incoming RIS messages and
// maintains peer state in the database.
import ( import (
"strconv" "strconv"
@@ -92,8 +93,10 @@ func (h *PeerHandler) QueueCapacity() int {
func (h *PeerHandler) HandleMessage(msg *ristypes.RISMessage) { func (h *PeerHandler) HandleMessage(msg *ristypes.RISMessage) {
// Parse peer ASN from string // Parse peer ASN from string
peerASN := 0 peerASN := 0
if msg.PeerASN != "" { if msg.PeerASN != "" {
if asn, err := strconv.Atoi(msg.PeerASN); err == nil { asn, err := strconv.Atoi(msg.PeerASN)
if err == nil {
peerASN = asn peerASN = asn
} }
} }
@@ -115,9 +118,16 @@ func (h *PeerHandler) HandleMessage(msg *ristypes.RISMessage) {
} }
} }
// Stop gracefully stops the handler and flushes remaining batches
func (h *PeerHandler) Stop() {
close(h.stopCh)
h.wg.Wait()
}
// flushLoop runs in a goroutine and periodically flushes batches // flushLoop runs in a goroutine and periodically flushes batches
func (h *PeerHandler) flushLoop() { func (h *PeerHandler) flushLoop() {
defer h.wg.Done() defer h.wg.Done()
ticker := time.NewTicker(peerBatchTimeout) ticker := time.NewTicker(peerBatchTimeout)
defer ticker.Stop() defer ticker.Stop()
@@ -140,7 +150,8 @@ func (h *PeerHandler) flushLoop() {
} }
} }
// flushBatchLocked flushes the peer batch to the database (must be called with mutex held) // flushBatchLocked flushes the peer batch to the database (must be called
// with mutex held)
func (h *PeerHandler) flushBatchLocked() { func (h *PeerHandler) flushBatchLocked() {
if len(h.peerBatch) == 0 { if len(h.peerBatch) == 0 {
return return
@@ -149,7 +160,8 @@ func (h *PeerHandler) flushBatchLocked() {
// Deduplicate by peer IP, keeping the latest update for each peer // Deduplicate by peer IP, keeping the latest update for each peer
peerMap := make(map[string]peerUpdate) peerMap := make(map[string]peerUpdate)
for _, update := range h.peerBatch { for _, update := range h.peerBatch {
if existing, ok := peerMap[update.peerIP]; !ok || update.timestamp.After(existing.timestamp) { existing, ok := peerMap[update.peerIP]
if !ok || update.timestamp.After(existing.timestamp) {
peerMap[update.peerIP] = update peerMap[update.peerIP] = update
} }
} }
@@ -166,17 +178,13 @@ func (h *PeerHandler) flushBatchLocked() {
} }
// Process all peers in a single batch transaction // Process all peers in a single batch transaction
if err := h.db.UpdatePeerBatch(dbPeerMap); err != nil { err := h.db.UpdatePeerBatch(dbPeerMap)
h.logger.Error("Failed to process peer batch", "error", err, "count", len(dbPeerMap)) if err != nil {
h.logger.Error("Failed to process peer batch",
"error", err, "count", len(dbPeerMap))
} }
// Clear batch // Clear batch
h.peerBatch = h.peerBatch[:0] h.peerBatch = h.peerBatch[:0]
h.lastFlush = time.Now() h.lastFlush = time.Now()
} }
// Stop gracefully stops the handler and flushes remaining batches
func (h *PeerHandler) Stop() {
close(h.stopCh)
h.wg.Wait()
}
internal/routewatch/peeringhandler.go
+24 -19
View File
@@ -72,7 +72,7 @@ func NewPeeringHandler(db database.Store, logger *logger.Logger) *PeeringHandler
// the AS path information needed to extract peering relationships. // the AS path information needed to extract peering relationships.
func (h *PeeringHandler) WantsMessage(messageType string) bool { func (h *PeeringHandler) WantsMessage(messageType string) bool {
// We only care about UPDATE messages that have AS paths // We only care about UPDATE messages that have AS paths
return messageType == "UPDATE" return messageType == updateMessageType
} }
// QueueCapacity returns the buffer size for the handler's message queue. // QueueCapacity returns the buffer size for the handler's message queue.
@@ -101,6 +101,7 @@ func (h *PeeringHandler) HandleMessage(msg *ristypes.RISMessage) {
return return
} }
key := string(pathJSON) key := string(pathJSON)
h.mu.Lock() h.mu.Lock()
@@ -116,6 +117,23 @@ func (h *PeeringHandler) HandleMessage(msg *ristypes.RISMessage) {
h.mu.Unlock() h.mu.Unlock()
} }
// ProcessPeeringsNow triggers immediate processing of all accumulated AS
// paths into peering records. This bypasses the normal periodic processing
// schedule and is primarily intended for testing purposes.
func (h *PeeringHandler) ProcessPeeringsNow() {
h.processPeerings()
}
// Stop gracefully shuts down the handler by signaling the background
// goroutine to stop and performing a final synchronous processing of
// any remaining AS paths. This ensures no peering data is lost during
// shutdown.
func (h *PeeringHandler) Stop() {
close(h.stopCh)
// Process any remaining peerings synchronously
h.processPeerings()
}
// processLoop runs periodically to process AS paths into peerings // processLoop runs periodically to process AS paths into peerings
func (h *PeeringHandler) processLoop() { func (h *PeeringHandler) processLoop() {
ticker := time.NewTicker(peeringProcessInterval) ticker := time.NewTicker(peeringProcessInterval)
@@ -134,13 +152,6 @@ func (h *PeeringHandler) processLoop() {
} }
} }
// ProcessPeeringsNow triggers immediate processing of all accumulated AS
// paths into peering records. This bypasses the normal periodic processing
// schedule and is primarily intended for testing purposes.
func (h *PeeringHandler) ProcessPeeringsNow() {
h.processPeerings()
}
// processPeerings extracts peerings from AS paths and writes to database // processPeerings extracts peerings from AS paths and writes to database
func (h *PeeringHandler) processPeerings() { func (h *PeeringHandler) processPeerings() {
// Take the accumulated paths and replace the map with a fresh empty one // Take the accumulated paths and replace the map with a fresh empty one
@@ -160,11 +171,14 @@ func (h *PeeringHandler) processPeerings() {
type peeringKey struct { type peeringKey struct {
low, high int low, high int
} }
peerings := make(map[peeringKey]time.Time) peerings := make(map[peeringKey]time.Time)
for pathJSON, timestamp := range paths { for pathJSON, timestamp := range paths {
var path []int var path []int
if err := json.Unmarshal([]byte(pathJSON), &path); err != nil {
err := json.Unmarshal([]byte(pathJSON), &path)
if err != nil {
h.logger.Error("Failed to decode AS path", "error", err) h.logger.Error("Failed to decode AS path", "error", err)
continue continue
@@ -197,6 +211,7 @@ func (h *PeeringHandler) processPeerings() {
// Record peerings in database // Record peerings in database
start := time.Now() start := time.Now()
successCount := 0 successCount := 0
for key, ts := range peerings { for key, ts := range peerings {
err := h.db.RecordPeering(key.low, key.high, ts) err := h.db.RecordPeering(key.low, key.high, ts)
if err != nil { if err != nil {
@@ -218,13 +233,3 @@ func (h *PeeringHandler) processPeerings() {
"duration", time.Since(start), "duration", time.Since(start),
) )
} }
// Stop gracefully shuts down the handler by signaling the background
// goroutine to stop and performing a final synchronous processing of
// any remaining AS paths. This ensures no peering data is lost during
// shutdown.
func (h *PeeringHandler) Stop() {
close(h.stopCh)
// Process any remaining peerings synchronously
h.processPeerings()
}
internal/routewatch/peeringhandler_test.go → internal/routewatch/peeringhandler_internal_test.go
+14 -2
View File
@@ -53,7 +53,8 @@ func newTestHandler(db database.Store) *PeeringHandler {
} }
} }
func snapshot(h *PeeringHandler) (tracked, dropped int) { // snapshot returns how many paths the handler tracks and how many it dropped.
func snapshot(h *PeeringHandler) (int, int) {
h.mu.Lock() h.mu.Lock()
defer h.mu.Unlock() defer h.mu.Unlock()
@@ -82,6 +83,8 @@ func handle(h *PeeringHandler, ts time.Time, asns ...int) {
// empties the path map (the swap) and that a path seen again after a run is // empties the path map (the swap) and that a path seen again after a run is
// recorded in the next run too. // recorded in the next run too.
func TestPeeringHandlerProcessesEachRunAndEmpties(t *testing.T) { func TestPeeringHandlerProcessesEachRunAndEmpties(t *testing.T) {
t.Parallel()
store := &recordingStore{mockStore: newMockStore()} store := &recordingStore{mockStore: newMockStore()}
h := newTestHandler(store) h := newTestHandler(store)
@@ -94,6 +97,7 @@ func TestPeeringHandlerProcessesEachRunAndEmpties(t *testing.T) {
if tracked, _ := snapshot(h); tracked != 0 { if tracked, _ := snapshot(h); tracked != 0 {
t.Fatalf("map not empty after first run: %d paths remain", tracked) t.Fatalf("map not empty after first run: %d paths remain", tracked)
} }
if got := store.callCount(); got != 1 { if got := store.callCount(); got != 1 {
t.Fatalf("want 1 RecordPeering call after first run, got %d", got) t.Fatalf("want 1 RecordPeering call after first run, got %d", got)
} }
@@ -105,6 +109,7 @@ func TestPeeringHandlerProcessesEachRunAndEmpties(t *testing.T) {
if tracked, _ := snapshot(h); tracked != 0 { if tracked, _ := snapshot(h); tracked != 0 {
t.Fatalf("map not empty after second run: %d paths remain", tracked) t.Fatalf("map not empty after second run: %d paths remain", tracked)
} }
if got := store.callCount(); got != 2 { if got := store.callCount(); got != 2 {
t.Fatalf("want 2 RecordPeering calls after second run, got %d", got) t.Fatalf("want 2 RecordPeering calls after second run, got %d", got)
} }
@@ -114,6 +119,8 @@ func TestPeeringHandlerProcessesEachRunAndEmpties(t *testing.T) {
// and counts them, while a path already tracked is refreshed rather than // and counts them, while a path already tracked is refreshed rather than
// dropped. // dropped.
func TestPeeringHandlerCapDropsAndCounts(t *testing.T) { func TestPeeringHandlerCapDropsAndCounts(t *testing.T) {
t.Parallel()
store := &recordingStore{mockStore: newMockStore()} store := &recordingStore{mockStore: newMockStore()}
h := newTestHandler(store) h := newTestHandler(store)
@@ -126,6 +133,7 @@ func TestPeeringHandlerCapDropsAndCounts(t *testing.T) {
h.mu.Lock() h.mu.Lock()
h.asPaths[existing] = now h.asPaths[existing] = now
for i := 0; len(h.asPaths) < maxTrackedPaths; i++ { for i := 0; len(h.asPaths) < maxTrackedPaths; i++ {
h.asPaths[strconv.Itoa(i)] = now h.asPaths[strconv.Itoa(i)] = now
} }
@@ -138,6 +146,7 @@ func TestPeeringHandlerCapDropsAndCounts(t *testing.T) {
if tracked != maxTrackedPaths { if tracked != maxTrackedPaths {
t.Fatalf("want map size %d after drop, got %d", maxTrackedPaths, tracked) t.Fatalf("want map size %d after drop, got %d", maxTrackedPaths, tracked)
} }
if dropped != 1 { if dropped != 1 {
t.Fatalf("want dropped count 1, got %d", dropped) t.Fatalf("want dropped count 1, got %d", dropped)
} }
@@ -150,6 +159,7 @@ func TestPeeringHandlerCapDropsAndCounts(t *testing.T) {
if tracked != maxTrackedPaths { if tracked != maxTrackedPaths {
t.Fatalf("want map size %d after refresh, got %d", maxTrackedPaths, tracked) t.Fatalf("want map size %d after refresh, got %d", maxTrackedPaths, tracked)
} }
if dropped != 1 { if dropped != 1 {
t.Fatalf("want dropped count still 1 after refresh, got %d", dropped) t.Fatalf("want dropped count still 1 after refresh, got %d", dropped)
} }
@@ -157,7 +167,9 @@ func TestPeeringHandlerCapDropsAndCounts(t *testing.T) {
h.mu.Lock() h.mu.Lock()
gotTS := h.asPaths[existing] gotTS := h.asPaths[existing]
h.mu.Unlock() h.mu.Unlock()
if !gotTS.Equal(refreshed) { if !gotTS.Equal(refreshed) {
t.Fatalf("existing path timestamp not refreshed: want %v, got %v", refreshed, gotTS) t.Fatalf("existing path timestamp not refreshed: want %v, got %v",
refreshed, gotTS)
} }
} }
internal/routewatch/prefixhandler.go
+112 -249
View File
@@ -2,6 +2,7 @@ package routewatch
import ( import (
"net" "net"
"net/netip"
"strings" "strings"
"sync" "sync"
"time" "time"
@@ -32,8 +33,8 @@ const (
ipv6Version = 6 ipv6Version = 6
) )
// PrefixHandler tracks BGP prefixes and maintains a live routing table in the database. // PrefixHandler tracks BGP prefixes and maintains a live routing table in the
// Routes are added on announcement and deleted on withdrawal. // database. Routes are added on announcement and deleted on withdrawal.
type PrefixHandler struct { type PrefixHandler struct {
db database.Store db database.Store
logger *logger.Logger logger *logger.Logger
@@ -78,10 +79,11 @@ func (h *PrefixHandler) SetMetricsTracker(metrics *metrics.Tracker) {
h.metrics = metrics h.metrics = metrics
} }
// WantsMessage returns true if this handler wants to process messages of the given type // WantsMessage returns true if this handler wants to process messages of the
// given type
func (h *PrefixHandler) WantsMessage(messageType string) bool { func (h *PrefixHandler) WantsMessage(messageType string) bool {
// We only care about UPDATE messages for the routing table // We only care about UPDATE messages for the routing table
return messageType == "UPDATE" return messageType == updateMessageType
} }
// QueueCapacity returns the desired queue capacity for this handler // QueueCapacity returns the desired queue capacity for this handler
@@ -104,11 +106,28 @@ func (h *PrefixHandler) HandleMessage(msg *ristypes.RISMessage) {
h.mu.Lock() h.mu.Lock()
defer h.mu.Unlock() defer h.mu.Unlock()
// Process announcements // Withdrawals are queued before announcements, because the batch keeps
// each peer's last update for a prefix: a message that lists a prefix as
// both withdrawn and announced keeps the route (RFC 4271 section 4.3).
for _, prefix := range msg.Withdrawals {
h.batch = append(h.batch, prefixUpdate{
prefix: canonicalPrefix(prefix),
originASN: originASN, // Use the originASN from path if available
peer: msg.Peer,
messageType: "withdrawal",
timestamp: timestamp,
path: msg.Path,
})
// Record withdrawal in metrics
if h.metrics != nil {
h.metrics.RecordWithdrawal()
}
}
for _, announcement := range msg.Announcements { for _, announcement := range msg.Announcements {
for _, prefix := range announcement.Prefixes { for _, prefix := range announcement.Prefixes {
h.batch = append(h.batch, prefixUpdate{ h.batch = append(h.batch, prefixUpdate{
prefix: prefix, prefix: canonicalPrefix(prefix),
originASN: originASN, originASN: originASN,
peer: msg.Peer, peer: msg.Peer,
messageType: "announcement", messageType: "announcement",
@@ -122,31 +141,22 @@ func (h *PrefixHandler) HandleMessage(msg *ristypes.RISMessage) {
} }
} }
// Process withdrawals
for _, prefix := range msg.Withdrawals {
h.batch = append(h.batch, prefixUpdate{
prefix: prefix,
originASN: originASN, // Use the originASN from path if available
peer: msg.Peer,
messageType: "withdrawal",
timestamp: timestamp,
path: msg.Path,
})
// Record withdrawal in metrics
if h.metrics != nil {
h.metrics.RecordWithdrawal()
}
}
// Check if we need to flush // Check if we need to flush
if len(h.batch) >= prefixBatchSize { if len(h.batch) >= prefixBatchSize {
h.flushBatchLocked() h.flushBatchLocked()
} }
} }
// Stop gracefully stops the handler and flushes remaining batches
func (h *PrefixHandler) Stop() {
close(h.stopCh)
h.wg.Wait()
}
// flushLoop runs in a goroutine and periodically flushes batches // flushLoop runs in a goroutine and periodically flushes batches
func (h *PrefixHandler) flushLoop() { func (h *PrefixHandler) flushLoop() {
defer h.wg.Done() defer h.wg.Done()
ticker := time.NewTicker(prefixBatchTimeout) ticker := time.NewTicker(prefixBatchTimeout)
defer ticker.Stop() defer ticker.Stop()
@@ -169,7 +179,8 @@ func (h *PrefixHandler) flushLoop() {
} }
} }
// flushBatchLocked flushes the prefix batch to the database (must be called with mutex held) // flushBatchLocked flushes the prefix batch to the database (must be called
// with mutex held)
func (h *PrefixHandler) flushBatchLocked() { func (h *PrefixHandler) flushBatchLocked() {
if len(h.batch) == 0 { if len(h.batch) == 0 {
return return
@@ -178,26 +189,31 @@ func (h *PrefixHandler) flushBatchLocked() {
startTime := time.Now() startTime := time.Now()
batchSize := len(h.batch) batchSize := len(h.batch)
// Group updates by prefix to deduplicate // For each prefix and peer, keep only the last update the feed delivered.
// For each prefix, keep the latest update // Updates from different peers for the same prefix are all kept. Arrival
prefixMap := make(map[string]prefixUpdate) // order decides, not the timestamp: timestamps are whole seconds, so
// updates within one second carry the same one.
type prefixAndPeer struct{ prefix, peer string }
lastUpdates := make(map[prefixAndPeer]prefixUpdate)
for _, update := range h.batch { for _, update := range h.batch {
key := update.prefix lastUpdates[prefixAndPeer{update.prefix, update.peer}] = update
if existing, ok := prefixMap[key]; !ok || update.timestamp.After(existing.timestamp) {
prefixMap[key] = update
}
} }
// Collect routes to upsert and delete // Collect routes to upsert and delete
var routesToUpsert []*database.LiveRoute var (
var routesToDelete []database.LiveRouteDeletion routesToUpsert []*database.LiveRoute
routesToDelete []database.LiveRouteDeletion
)
// Collect unique prefixes to update // Collect unique prefixes to update
prefixesToUpdate := make(map[string]time.Time) prefixesToUpdate := make(map[string]time.Time)
for _, update := range prefixMap { for _, update := range lastUpdates {
// Track prefix for both announcements and withdrawals // Track prefix for both announcements and withdrawals
if _, exists := prefixesToUpdate[update.prefix]; !exists || update.timestamp.After(prefixesToUpdate[update.prefix]) { seen, exists := prefixesToUpdate[update.prefix]
if !exists || update.timestamp.After(seen) {
prefixesToUpdate[update.prefix] = update.timestamp prefixesToUpdate[update.prefix] = update.timestamp
} }
@@ -211,7 +227,8 @@ func (h *PrefixHandler) flushBatchLocked() {
// Parse CIDR to get IP version // Parse CIDR to get IP version
_, ipVersion, err := parseCIDR(update.prefix) _, ipVersion, err := parseCIDR(update.prefix)
if err != nil { if err != nil {
h.logger.Error("Failed to parse CIDR for withdrawal", "prefix", update.prefix, "error", err) h.logger.Error("Failed to parse CIDR for withdrawal",
"prefix", update.prefix, "error", err)
continue continue
} }
@@ -226,35 +243,12 @@ func (h *PrefixHandler) flushBatchLocked() {
} }
} }
// Process batch operations successCount := h.writeBatch(routesToUpsert, routesToDelete, prefixesToUpdate)
successCount := 0
if len(routesToUpsert) > 0 {
if err := h.db.UpsertLiveRouteBatch(routesToUpsert); err != nil {
h.logger.Error("Failed to upsert route batch", "error", err, "count", len(routesToUpsert))
} else {
successCount += len(routesToUpsert)
}
}
if len(routesToDelete) > 0 {
if err := h.db.DeleteLiveRouteBatch(routesToDelete); err != nil {
h.logger.Error("Failed to delete route batch", "error", err, "count", len(routesToDelete))
} else {
successCount += len(routesToDelete)
}
}
// Update prefix tables
if len(prefixesToUpdate) > 0 {
if err := h.db.UpdatePrefixesBatch(prefixesToUpdate); err != nil {
h.logger.Error("Failed to update prefix batch", "error", err, "count", len(prefixesToUpdate))
}
}
elapsed := time.Since(startTime) elapsed := time.Since(startTime)
h.logger.Debug("Flushed prefix batch", h.logger.Debug("Flushed prefix batch",
"batch_size", batchSize, "batch_size", batchSize,
"unique_prefixes", len(prefixMap), "unique_prefixes", len(prefixesToUpdate),
"success", successCount, "success", successCount,
"duration_ms", elapsed.Milliseconds(), "duration_ms", elapsed.Milliseconds(),
) )
@@ -264,8 +258,64 @@ func (h *PrefixHandler) flushBatchLocked() {
h.lastFlush = time.Now() h.lastFlush = time.Now()
} }
// parseCIDR extracts the mask length and IP version from a prefix string // writeBatch writes the collected route upserts, route deletions and prefix
func parseCIDR(prefix string) (maskLength int, ipVersion int, err error) { // updates to the database and returns how many routes it wrote.
func (h *PrefixHandler) writeBatch(
routesToUpsert []*database.LiveRoute,
routesToDelete []database.LiveRouteDeletion,
prefixesToUpdate map[string]time.Time,
) int {
successCount := 0
if len(routesToUpsert) > 0 {
err := h.db.UpsertLiveRouteBatch(routesToUpsert)
if err != nil {
h.logger.Error("Failed to upsert route batch",
"error", err, "count", len(routesToUpsert))
} else {
successCount += len(routesToUpsert)
}
}
if len(routesToDelete) > 0 {
err := h.db.DeleteLiveRouteBatch(routesToDelete)
if err != nil {
h.logger.Error("Failed to delete route batch",
"error", err, "count", len(routesToDelete))
} else {
successCount += len(routesToDelete)
}
}
// Update prefix tables
if len(prefixesToUpdate) > 0 {
err := h.db.UpdatePrefixesBatch(prefixesToUpdate)
if err != nil {
h.logger.Error("Failed to update prefix batch",
"error", err, "count", len(prefixesToUpdate))
}
}
return successCount
}
// canonicalPrefix returns prefix in the text form net/netip prints, the form
// the IP lookup builds when it looks a prefix up. The feed sends IPv6
// announcements compressed ("2001:db8::/32") but IPv6 withdrawals uncompressed
// ("2001:db8:0:0:0:0:0:0/32"); stored as received, a withdrawal would not match
// the route its announcement stored. A prefix that does not parse is returned
// unchanged, and the batch flush reports it.
func canonicalPrefix(prefix string) string {
p, err := netip.ParsePrefix(prefix)
if err != nil {
return prefix
}
return p.Masked().String()
}
// parseCIDR returns the mask length and the IP version of a prefix string
func parseCIDR(prefix string) (int, int, error) {
_, ipNet, err := net.ParseCIDR(prefix) _, ipNet, err := net.ParseCIDR(prefix)
if err != nil { if err != nil {
return 0, 0, err return 0, 0, err
@@ -279,64 +329,6 @@ func parseCIDR(prefix string) (maskLength int, ipVersion int, err error) {
return ones, ipv4Version, nil return ones, ipv4Version, nil
} }
// processAnnouncement handles storing an announcement in the database
// nolint:unused // kept for potential future use
func (h *PrefixHandler) processAnnouncement(_ *database.Prefix, update prefixUpdate) {
// Parse CIDR to get mask length
maskLength, ipVersion, err := parseCIDR(update.prefix)
if err != nil {
h.logger.Error("Failed to parse CIDR",
"prefix", update.prefix,
"error", err,
)
return
}
// Track route update metrics
if h.metrics != nil {
if ipVersion == ipv4Version {
h.metrics.RecordIPv4Update()
} else {
h.metrics.RecordIPv6Update()
}
}
// Create live route record
liveRoute := &database.LiveRoute{
ID: uuid.New(),
Prefix: update.prefix,
MaskLength: maskLength,
IPVersion: ipVersion,
OriginASN: update.originASN,
PeerIP: update.peer,
ASPath: update.path,
NextHop: update.peer, // Using peer as next hop
LastUpdated: update.timestamp,
}
// For IPv4, calculate the IP range
if ipVersion == ipv4Version {
start, end, err := database.CalculateIPv4Range(update.prefix)
if err == nil {
liveRoute.V4IPStart = &start
liveRoute.V4IPEnd = &end
} else {
h.logger.Error("Failed to calculate IPv4 range",
"prefix", update.prefix,
"error", err,
)
}
}
if err := h.db.UpsertLiveRoute(liveRoute); err != nil {
h.logger.Error("Failed to upsert live route",
"prefix", update.prefix,
"error", err,
)
}
}
// createLiveRoute creates a LiveRoute from a prefix update // createLiveRoute creates a LiveRoute from a prefix update
func (h *PrefixHandler) createLiveRoute(update prefixUpdate) *database.LiveRoute { func (h *PrefixHandler) createLiveRoute(update prefixUpdate) *database.LiveRoute {
// Parse CIDR to get mask length // Parse CIDR to get mask length
@@ -372,134 +364,5 @@ func (h *PrefixHandler) createLiveRoute(update prefixUpdate) *database.LiveRoute
LastUpdated: update.timestamp, LastUpdated: update.timestamp,
} }
// For IPv4, calculate the IP range
if ipVersion == ipv4Version {
start, end, err := database.CalculateIPv4Range(update.prefix)
if err == nil {
liveRoute.V4IPStart = &start
liveRoute.V4IPEnd = &end
} else {
h.logger.Error("Failed to calculate IPv4 range",
"prefix", update.prefix,
"error", err,
)
}
}
return liveRoute return liveRoute
} }
// processAnnouncementDirect handles storing an announcement directly without prefix table
// nolint:unused // kept for potential future use
func (h *PrefixHandler) processAnnouncementDirect(update prefixUpdate) {
// Parse CIDR to get mask length
maskLength, ipVersion, err := parseCIDR(update.prefix)
if err != nil {
h.logger.Error("Failed to parse CIDR",
"prefix", update.prefix,
"error", err,
)
return
}
// Track route update metrics
if h.metrics != nil {
if ipVersion == ipv4Version {
h.metrics.RecordIPv4Update()
} else {
h.metrics.RecordIPv6Update()
}
}
// Create live route record
liveRoute := &database.LiveRoute{
ID: uuid.New(),
Prefix: update.prefix,
MaskLength: maskLength,
IPVersion: ipVersion,
OriginASN: update.originASN,
PeerIP: update.peer,
ASPath: update.path,
NextHop: update.peer, // Using peer as next hop
LastUpdated: update.timestamp,
}
// For IPv4, calculate the IP range
if ipVersion == ipv4Version {
start, end, err := database.CalculateIPv4Range(update.prefix)
if err == nil {
liveRoute.V4IPStart = &start
liveRoute.V4IPEnd = &end
} else {
h.logger.Error("Failed to calculate IPv4 range",
"prefix", update.prefix,
"error", err,
)
}
}
if err := h.db.UpsertLiveRoute(liveRoute); err != nil {
h.logger.Error("Failed to upsert live route",
"prefix", update.prefix,
"error", err,
)
}
}
// processWithdrawalDirect handles removing a route directly without prefix table
// nolint:unused // kept for potential future use
func (h *PrefixHandler) processWithdrawalDirect(update prefixUpdate) {
// For withdrawals, we need to delete the route from live_routes
if update.originASN > 0 {
if err := h.db.DeleteLiveRoute(update.prefix, update.originASN, update.peer); err != nil {
h.logger.Error("Failed to delete live route",
"prefix", update.prefix,
"origin_asn", update.originASN,
"peer", update.peer,
"error", err,
)
}
} else {
// If no origin ASN, just delete all routes for this prefix from this peer
if err := h.db.DeleteLiveRoute(update.prefix, 0, update.peer); err != nil {
h.logger.Error("Failed to delete live route (no origin ASN)",
"prefix", update.prefix,
"peer", update.peer,
"error", err,
)
}
}
}
// processWithdrawal handles removing a route from the live routing table
// nolint:unused // kept for potential future use
func (h *PrefixHandler) processWithdrawal(_ *database.Prefix, update prefixUpdate) {
// For withdrawals, we need to delete the route from live_routes
// Since we have the origin ASN from the update, we can delete the specific route
if update.originASN > 0 {
if err := h.db.DeleteLiveRoute(update.prefix, update.originASN, update.peer); err != nil {
h.logger.Error("Failed to delete live route",
"prefix", update.prefix,
"origin_asn", update.originASN,
"peer", update.peer,
"error", err,
)
}
} else {
// If no origin ASN, just delete all routes for this prefix from this peer
if err := h.db.DeleteLiveRoute(update.prefix, 0, update.peer); err != nil {
h.logger.Error("Failed to delete live route (no origin ASN)",
"prefix", update.prefix,
"peer", update.peer,
"error", err,
)
}
}
}
// Stop gracefully stops the handler and flushes remaining batches
func (h *PrefixHandler) Stop() {
close(h.stopCh)
h.wg.Wait()
}
internal/routewatch/prefixhandler_internal_test.go
+268
View File
@@ -0,0 +1,268 @@
package routewatch
import (
"context"
"errors"
"slices"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/database"
"git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/ristypes"
)
const (
testPeerIP = "2001:db8:ffff::1"
testV4Prefix = "192.0.2.0/24"
)
// TestPrefixHandlerStoresPrefixesTheIPLookupFinds runs announcements and
// withdrawals through the prefix handler into a real database and looks the
// addresses up. The prefixes are written the way the feed sends them: IPv6
// announcements compressed, IPv6 withdrawals uncompressed. The withdrawal must
// remove the route the announcement stored, and the IP lookup must find the
// stored prefix.
func TestPrefixHandlerStoresPrefixesTheIPLookupFinds(t *testing.T) {
t.Parallel()
db, err := database.New(&config.Config{StateDir: t.TempDir()}, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
defer func() { _ = db.Close() }()
// Built without NewPrefixHandler's flush timer, so the test flushes itself.
h := &PrefixHandler{db: db, logger: logger.New()}
flush := func() {
h.mu.Lock()
defer h.mu.Unlock()
h.flushBatchLocked()
}
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
h.HandleMessage(&ristypes.RISMessage{
Peer: testPeerIP,
Path: ristypes.ASPath{testASNA, testASNB},
ParsedTimestamp: ts,
Announcements: []ristypes.RISAnnouncement{{
NextHop: testPeerIP,
Prefixes: []string{"2001:db8:1::/48", testV4Prefix},
}},
})
flush()
for ip, want := range map[string]string{
"2001:db8:1::1": "2001:db8:1::/48",
"192.0.2.1": testV4Prefix,
} {
info, err := db.GetIPInfo(ip)
if err != nil {
t.Fatalf("GetIPInfo(%s) after announcement: %v", ip, err)
}
if info.Netblock != want || info.ASN != testASNB {
t.Errorf("GetIPInfo(%s) = %s AS%d, want %s AS%d",
ip, info.Netblock, info.ASN, want, testASNB)
}
}
h.HandleMessage(&ristypes.RISMessage{
Peer: testPeerIP,
ParsedTimestamp: ts.Add(time.Minute),
Withdrawals: []string{"2001:db8:1:0:0:0:0:0/48", testV4Prefix},
})
flush()
for _, ip := range []string{"2001:db8:1::1", "192.0.2.1"} {
info, err := db.GetIPInfo(ip)
if !errors.Is(err, database.ErrNoRoute) {
t.Errorf("GetIPInfo(%s) after withdrawal = %+v, %v; want ErrNoRoute",
ip, info, err)
}
}
}
// TestPrefixHandlerAppliesEachPeersLastUpdateInABatch runs updates for one
// prefix through the prefix handler into a real database. All updates carry the
// same timestamp, as updates within one second of the feed do. Within a batch,
// the last update each peer sent for the prefix decides whether that peer has a
// live route for it, and updates from different peers are all applied. The
// in-memory route count and prefix distribution must match the route table.
func TestPrefixHandlerAppliesEachPeersLastUpdateInABatch(t *testing.T) {
t.Parallel()
const (
prefix = "198.51.100.0/24"
prefixMaskLength = 24
otherPeerIP = "2001:db8:ffff::2"
)
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
announce := func(peer string) *ristypes.RISMessage {
return &ristypes.RISMessage{
Peer: peer,
Path: ristypes.ASPath{testASNA, testASNB},
ParsedTimestamp: ts,
Announcements: []ristypes.RISAnnouncement{
{NextHop: peer, Prefixes: []string{prefix}},
},
}
}
withdraw := func(peer string) *ristypes.RISMessage {
return &ristypes.RISMessage{
Peer: peer, ParsedTimestamp: ts, Withdrawals: []string{prefix},
}
}
announceAndWithdraw := func(peer string) *ristypes.RISMessage {
msg := announce(peer)
msg.Withdrawals = []string{prefix}
return msg
}
tests := []struct {
name string
// Each inner list is handled and then flushed as one batch.
batches [][]*ristypes.RISMessage
// The peers left with a live route for the prefix, by peer IP.
wantPeers []string
}{
{
name: "two peers announce the prefix",
batches: [][]*ristypes.RISMessage{{announce(testPeerIP), announce(otherPeerIP)}},
wantPeers: []string{testPeerIP, otherPeerIP},
},
{
name: "two peers withdraw the prefix",
batches: [][]*ristypes.RISMessage{
{announce(testPeerIP)},
{announce(otherPeerIP)},
{withdraw(testPeerIP), withdraw(otherPeerIP)},
},
},
{
name: "a peer announces and then withdraws the prefix",
batches: [][]*ristypes.RISMessage{{announce(testPeerIP), withdraw(testPeerIP)}},
},
{
name: "a peer withdraws and then re-announces the prefix",
batches: [][]*ristypes.RISMessage{
{announce(testPeerIP)},
{withdraw(testPeerIP), announce(testPeerIP)},
},
wantPeers: []string{testPeerIP},
},
{
// RFC 4271 section 4.3: the announcement counts, not the withdrawal.
name: "one message announces and withdraws the prefix",
batches: [][]*ristypes.RISMessage{
{announce(testPeerIP)},
{announceAndWithdraw(testPeerIP)},
},
wantPeers: []string{testPeerIP},
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
t.Parallel()
db := handleBatches(t, tt.batches)
checkPrefixRoutes(t, db, prefix, prefixMaskLength, tt.wantPeers)
})
}
}
// handleBatches runs each batch of messages through a prefix handler on a new
// database, flushing after each batch, and returns the database. The database
// is closed when the test ends.
func handleBatches(
t *testing.T, batches [][]*ristypes.RISMessage,
) *database.Database {
t.Helper()
db, err := database.New(&config.Config{StateDir: t.TempDir()}, logger.New())
if err != nil {
t.Fatalf("failed to create database: %v", err)
}
t.Cleanup(func() { _ = db.Close() })
// Built without NewPrefixHandler's flush timer, so the test flushes itself.
h := &PrefixHandler{db: db, logger: logger.New()}
for _, batch := range batches {
for _, msg := range batch {
h.HandleMessage(msg)
}
h.mu.Lock()
h.flushBatchLocked()
h.mu.Unlock()
}
return db
}
// checkPrefixRoutes checks that exactly wantPeers have a live route for the
// IPv4 prefix, and that the in-memory route count and prefix distribution
// match the route table.
func checkPrefixRoutes(
t *testing.T, db *database.Database, prefix string, maskLength int,
wantPeers []string,
) {
t.Helper()
ctx := context.Background()
routes, _, err := db.GetPrefixDetailsContext(ctx, prefix)
if err != nil && !errors.Is(err, database.ErrNoRoute) {
t.Fatalf("GetPrefixDetailsContext: %v", err)
}
gotPeers := make([]string, 0, len(routes))
for _, route := range routes {
gotPeers = append(gotPeers, route.PeerIP)
}
if !slices.Equal(gotPeers, wantPeers) {
t.Errorf("peers with a live route for %s = %v, want %v",
prefix, gotPeers, wantPeers)
}
stats, err := db.GetStatsContext(ctx)
if err != nil {
t.Fatalf("GetStatsContext: %v", err)
}
tableRoutesV4, _, err := db.GetLiveRouteCountsContext(ctx)
if err != nil {
t.Fatalf("GetLiveRouteCountsContext: %v", err)
}
if stats.IPv4Routes != len(wantPeers) || tableRoutesV4 != len(wantPeers) {
t.Errorf("IPv4 live routes: %d in memory, %d in the table, want %d",
stats.IPv4Routes, tableRoutesV4, len(wantPeers))
}
var wantDistribution []database.PrefixDistribution
if len(wantPeers) > 0 {
wantDistribution = []database.PrefixDistribution{
{MaskLength: maskLength, Count: 1},
}
}
queryDistributionV4, _, err := db.GetPrefixDistributionContext(ctx)
if err != nil {
t.Fatalf("GetPrefixDistributionContext: %v", err)
}
if !slices.Equal(stats.IPv4PrefixDistribution, wantDistribution) ||
!slices.Equal(queryDistributionV4, wantDistribution) {
t.Errorf("IPv4 prefix distribution: %v in memory, %v from the table, want %v",
stats.IPv4PrefixDistribution, queryDistributionV4, wantDistribution)
}
}
internal/server/handlers.go
+589 -544
View File
File diff suppressed because it is too large Load Diff
internal/server/handlers_internal_test.go
+358
View File
@@ -0,0 +1,358 @@
package server
import (
"context"
"encoding/json"
"errors"
"html"
"net/http"
"net/http/httptest"
"runtime"
"slices"
"strings"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/database"
"git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/metrics"
"git.eeqj.de/sneak/routewatch/internal/streamer"
"git.eeqj.de/sneak/routewatch/pkg/asinfo"
"github.com/google/uuid"
)
// The prefix and peer addresses the prefix page tests store routes for.
const (
testPrefixV4 = "192.0.2.0/24"
testPeerV4 = "198.51.100.1"
testPeerV6 = "2001:db8::1"
)
// blockingStatsDB embeds database.Store (left nil) and overrides only
// GetStatsContext, which blocks until release is closed. The stats handlers
// call it in a goroutine; every other Store method is unused on the timeout
// path and would panic if called.
type blockingStatsDB struct {
database.Store
release chan struct{}
}
func (d blockingStatsDB) GetStatsContext(_ context.Context) (database.Stats, error) {
<-d.release
return database.Stats{}, nil
}
// TestStatsHandlersDoNotLeakOnTimeout drives each stats handler repeatedly with
// a request whose context times out before the database responds, then releases
// the blocked queries and asserts the goroutine count returns to its starting
// value. Before the fix the per-request goroutine sent on an unbuffered channel
// that nothing received once the timeout won, so it blocked forever and every
// poll leaked one goroutine.
//
//nolint:paralleltest // counts goroutines, so no other test may run alongside
func TestStatsHandlersDoNotLeakOnTimeout(t *testing.T) {
release := make(chan struct{})
db := blockingStatsDB{release: release}
s := New(db, streamer.New(logger.New(), metrics.New()), logger.New(), &config.Config{})
handlers := map[string]http.HandlerFunc{
"status.json": s.handleStatusJSON(),
"stats": s.handleStats(),
}
baseline := settledGoroutineCount()
const (
iterations = 20
requestTimeout = 50 * time.Millisecond
)
for _, handler := range handlers {
for range iterations {
ctx, cancel := context.WithTimeout(context.Background(), requestTimeout)
req := httptest.NewRequestWithContext(ctx, http.MethodGet, "/", nil)
handler(httptest.NewRecorder(), req)
cancel()
}
}
// Let the blocked queries finish; with buffered channels each goroutine's
// send now succeeds and the goroutine exits.
close(release)
if !waitForGoroutines(baseline) {
t.Fatalf("goroutines did not return to baseline %d, got %d",
baseline, runtime.NumGoroutine())
}
}
// TestStatsHandlersAnswerFromMemory checks that both stats handlers answer 200
// with the live route counts and the prefix distribution while the database is
// closed, so that any query would fail: the request path reads them from
// memory. The prefix distribution query it used to run read every live route
// and, on a large database, took the whole 4-second deadline, so
// /api/v1/stats answered 500 (https://git.eeqj.de/sneak/routewatch/issues/30).
// The oldest and newest route times still come from one-row lookups at the ends
// of an index; with the database closed they are left out of the answer.
func TestStatsHandlersAnswerFromMemory(t *testing.T) {
t.Parallel()
db, err := database.New(&config.Config{StateDir: t.TempDir()}, logger.New())
if err != nil {
t.Fatalf("database.New: %v", err)
}
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
err = db.UpsertLiveRouteBatch([]*database.LiveRoute{
{
ID: uuid.New(), Prefix: "198.51.100.0/24", MaskLength: 24, IPVersion: 4,
OriginASN: 64500, PeerIP: "192.0.2.1", ASPath: []int{64500},
NextHop: "192.0.2.1", LastUpdated: ts,
},
{
ID: uuid.New(), Prefix: "2001:db8::/32", MaskLength: 32, IPVersion: 6,
OriginASN: 64501, PeerIP: testPeerV6, ASPath: []int{64501},
NextHop: testPeerV6, LastUpdated: ts,
},
})
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
err = db.Close()
if err != nil {
t.Fatalf("Close: %v", err)
}
s := New(db, streamer.New(logger.New(), metrics.New()), logger.New(), &config.Config{})
handlers := map[string]http.HandlerFunc{
"status.json": s.handleStatusJSON(),
"stats": s.handleStats(),
}
for name, handler := range handlers {
rec := httptest.NewRecorder()
handler(rec, httptest.NewRequestWithContext(t.Context(), http.MethodGet, "/", nil))
if rec.Code != http.StatusOK {
t.Errorf("%s: status %d, want %d; body %s",
name, rec.Code, http.StatusOK, rec.Body)
continue
}
//nolint:tagliatelle // field names are the existing JSON API
var body struct {
Data struct {
IPv4Routes int `json:"ipv4_routes"`
IPv6Routes int `json:"ipv6_routes"`
IPv4Distribution []database.PrefixDistribution `json:"ipv4_prefix_distribution"`
IPv6Distribution []database.PrefixDistribution `json:"ipv6_prefix_distribution"`
} `json:"data"`
}
err := json.Unmarshal(rec.Body.Bytes(), &body)
if err != nil {
t.Fatalf("%s: decoding the answer: %v", name, err)
}
if body.Data.IPv4Routes != 1 || body.Data.IPv6Routes != 1 {
t.Errorf("%s: routes = (v4 %d, v6 %d), want (1, 1)",
name, body.Data.IPv4Routes, body.Data.IPv6Routes)
}
wantV4 := []database.PrefixDistribution{{MaskLength: 24, Count: 1}}
if !slices.Equal(body.Data.IPv4Distribution, wantV4) {
t.Errorf("%s: IPv4 distribution = %v, want %v",
name, body.Data.IPv4Distribution, wantV4)
}
wantV6 := []database.PrefixDistribution{{MaskLength: 32, Count: 1}}
if !slices.Equal(body.Data.IPv6Distribution, wantV6) {
t.Errorf("%s: IPv6 distribution = %v, want %v",
name, body.Data.IPv6Distribution, wantV6)
}
}
}
// TestPrefixDetailHandlersParsePrefix checks that both prefix pages find a
// stored route whatever valid text form the URL writes its prefix in, and
// answer 400 for a prefix that does not parse. The tables store the form
// net/netip prints. Before the fix the URL text was looked up as typed, so
// "2001:DB8:0:0::/32" was not found although "2001:db8::/32" was stored, and
// "foo/24" answered 500 (https://git.eeqj.de/sneak/routewatch/issues/71).
func TestPrefixDetailHandlersParsePrefix(t *testing.T) {
t.Parallel()
db, err := database.New(&config.Config{StateDir: t.TempDir()}, logger.New())
if err != nil {
t.Fatalf("database.New: %v", err)
}
t.Cleanup(func() { _ = db.Close() })
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
err = db.UpsertLiveRouteBatch([]*database.LiveRoute{
{
ID: uuid.New(), Prefix: testPrefixV4, MaskLength: 24, IPVersion: 4,
OriginASN: 64500, PeerIP: testPeerV4, ASPath: []int{64500},
NextHop: testPeerV4, LastUpdated: ts,
},
{
ID: uuid.New(), Prefix: "2001:db8::/32", MaskLength: 32, IPVersion: 6,
OriginASN: 64501, PeerIP: testPeerV6, ASPath: []int{64501},
NextHop: testPeerV6, LastUpdated: ts,
},
})
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
s := New(db, streamer.New(logger.New(), metrics.New()), logger.New(), &config.Config{})
tests := []struct {
path string
want int
}{
{"/api/v1/prefix/foo/24", http.StatusBadRequest},
{"/prefix/foo/24", http.StatusBadRequest},
{"/api/v1/prefix/2001:DB8:0:0::/32", http.StatusOK},
{"/prefix/2001:DB8:0:0::/32", http.StatusOK},
{"/api/v1/prefix/192.0.2.1/24", http.StatusOK},
{"/prefix/192.0.2.1/24", http.StatusOK},
}
for _, tt := range tests {
rec := httptest.NewRecorder()
s.router.ServeHTTP(rec, httptest.NewRequestWithContext(
t.Context(), http.MethodGet, tt.path, nil))
if rec.Code != tt.want {
t.Errorf("%s: status %d, want %d; body %s",
tt.path, rec.Code, tt.want, rec.Body)
}
}
}
// errASDetailsLookupFailed is what failingASDetailsDB answers every AS details
// lookup with.
var errASDetailsLookupFailed = errors.New("AS details lookup failed")
// failingASDetailsDB wraps a real store and fails every AS details lookup.
type failingASDetailsDB struct {
database.Store
}
func (failingASDetailsDB) GetASDetailsContext(
_ context.Context, _ int,
) (*database.ASN, []database.LiveRoute, error) {
return nil, nil, errASDetailsLookupFailed
}
// TestPrefixDetailShowsOriginNames checks that the prefix page shows each
// origin AS's handle and description without looking up the AS's details,
// which also read every route the AS originates. Before the fix the page did
// that lookup once per origin, so with it failing the names were missing
// (https://git.eeqj.de/sneak/routewatch/issues/67). The AS path column shows
// handles too, so the test looks for the origin list's "(handle)" form.
func TestPrefixDetailShowsOriginNames(t *testing.T) {
t.Parallel()
db, err := database.New(&config.Config{StateDir: t.TempDir()}, logger.New())
if err != nil {
t.Fatalf("database.New: %v", err)
}
t.Cleanup(func() { _ = db.Close() })
ts := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
origins := []int{15169, 13335}
for _, asn := range origins {
_, err := db.GetOrCreateASN(asn, ts)
if err != nil {
t.Fatalf("GetOrCreateASN(%d): %v", asn, err)
}
}
err = db.UpsertLiveRouteBatch([]*database.LiveRoute{
{
ID: uuid.New(), Prefix: testPrefixV4, MaskLength: 24, IPVersion: 4,
OriginASN: origins[0], PeerIP: testPeerV4,
ASPath: []int{origins[0]}, NextHop: testPeerV4, LastUpdated: ts,
},
{
ID: uuid.New(), Prefix: testPrefixV4, MaskLength: 24, IPVersion: 4,
OriginASN: origins[1], PeerIP: "198.51.100.2",
ASPath: []int{origins[1]}, NextHop: "198.51.100.2", LastUpdated: ts,
},
})
if err != nil {
t.Fatalf("UpsertLiveRouteBatch: %v", err)
}
s := New(failingASDetailsDB{db}, streamer.New(logger.New(), metrics.New()),
logger.New(), &config.Config{})
rec := httptest.NewRecorder()
s.router.ServeHTTP(rec, httptest.NewRequestWithContext(
t.Context(), http.MethodGet, "/prefix/"+testPrefixV4, nil))
if rec.Code != http.StatusOK {
t.Fatalf("status %d, want %d; body %s", rec.Code, http.StatusOK, rec.Body)
}
page := rec.Body.String()
for _, asn := range origins {
info, ok := asinfo.Get(asn)
if !ok {
t.Fatalf("AS%d is not in the AS info data", asn)
}
want := "(" + html.EscapeString(info.Handle) + ")"
if !strings.Contains(page, want) {
t.Errorf("page does not show AS%d's handle %q", asn, want)
}
want = html.EscapeString(info.Description)
if !strings.Contains(page, want) {
t.Errorf("page does not show AS%d's description %q", asn, want)
}
}
}
// settledGoroutineCount lets transient goroutines finish, then reports the
// current count.
func settledGoroutineCount() int {
prev := runtime.NumGoroutine()
for range 20 {
time.Sleep(10 * time.Millisecond)
cur := runtime.NumGoroutine()
if cur == prev {
return cur
}
prev = cur
}
return prev
}
// waitForGoroutines waits until the goroutine count drops to target or below.
func waitForGoroutines(target int) bool {
for range 100 {
if runtime.NumGoroutine() <= target {
return true
}
time.Sleep(10 * time.Millisecond)
}
return false
}
internal/server/handlers_test.go
-99
View File
@@ -1,99 +0,0 @@
package server
import (
"context"
"net/http"
"net/http/httptest"
"runtime"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/database"
"git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/metrics"
"git.eeqj.de/sneak/routewatch/internal/streamer"
)
// blockingStatsDB embeds database.Store (left nil) and overrides only
// GetStatsContext, which blocks until release is closed. The stats handlers
// call it in a goroutine; every other Store method is unused on the timeout
// path and would panic if called.
type blockingStatsDB struct {
database.Store
release chan struct{}
}
func (d blockingStatsDB) GetStatsContext(_ context.Context) (database.Stats, error) {
<-d.release
return database.Stats{}, nil
}
// TestStatsHandlersDoNotLeakOnTimeout drives each stats handler repeatedly with
// a request whose context times out before the database responds, then releases
// the blocked queries and asserts the goroutine count returns to its starting
// value. Before the fix the per-request goroutine sent on an unbuffered channel
// that nothing received once the timeout won, so it blocked forever and every
// poll leaked one goroutine.
func TestStatsHandlersDoNotLeakOnTimeout(t *testing.T) {
release := make(chan struct{})
db := blockingStatsDB{release: release}
s := New(db, streamer.New(logger.New(), metrics.New()), logger.New())
handlers := map[string]http.HandlerFunc{
"status.json": s.handleStatusJSON(),
"stats": s.handleStats(),
}
baseline := settledGoroutineCount()
const (
iterations = 20
requestTimeout = 50 * time.Millisecond
)
for _, handler := range handlers {
for range iterations {
ctx, cancel := context.WithTimeout(context.Background(), requestTimeout)
req := httptest.NewRequest(http.MethodGet, "/", nil).WithContext(ctx)
handler(httptest.NewRecorder(), req)
cancel()
}
}
// Let the blocked queries finish; with buffered channels each goroutine's
// send now succeeds and the goroutine exits.
close(release)
if !waitForGoroutines(baseline) {
t.Fatalf("goroutines did not return to baseline %d, got %d",
baseline, runtime.NumGoroutine())
}
}
// settledGoroutineCount lets transient goroutines finish, then reports the
// current count.
func settledGoroutineCount() int {
prev := runtime.NumGoroutine()
for range 20 {
time.Sleep(10 * time.Millisecond)
cur := runtime.NumGoroutine()
if cur == prev {
return cur
}
prev = cur
}
return prev
}
// waitForGoroutines waits until the goroutine count drops to target or below.
func waitForGoroutines(target int) bool {
for range 100 {
if runtime.NumGoroutine() <= target {
return true
}
time.Sleep(10 * time.Millisecond)
}
return false
}
internal/server/middleware.go
+36 -21
View File
@@ -14,6 +14,7 @@ import (
// responseWriter wraps http.ResponseWriter to capture the response // responseWriter wraps http.ResponseWriter to capture the response
type responseWriter struct { type responseWriter struct {
http.ResponseWriter http.ResponseWriter
body *bytes.Buffer body *bytes.Buffer
statusCode int statusCode int
written bool written bool
@@ -86,8 +87,10 @@ func JSONResponseMiddleware(next http.Handler) http.Handler {
} }
// Parse the original response // Parse the original response
var originalResponse map[string]interface{} var originalResponse map[string]any
if err := json.Unmarshal(rw.body.Bytes(), &originalResponse); err != nil {
err := json.Unmarshal(rw.body.Bytes(), &originalResponse)
if err != nil {
// If we can't parse it, just write original // If we can't parse it, just write original
w.WriteHeader(rw.statusCode) w.WriteHeader(rw.statusCode)
_, _ = w.Write(rw.body.Bytes()) _, _ = w.Write(rw.body.Bytes())
@@ -96,7 +99,7 @@ func JSONResponseMiddleware(next http.Handler) http.Handler {
} }
// Add @meta field // Add @meta field
originalResponse["@meta"] = map[string]interface{}{ originalResponse["@meta"] = map[string]any{
"time_zone": "UTC", "time_zone": "UTC",
"api_version": 1, "api_version": 1,
"execution_time": fmt.Sprintf("%d ms", execTime.Milliseconds()), "execution_time": fmt.Sprintf("%d ms", execTime.Milliseconds()),
@@ -112,6 +115,7 @@ func JSONResponseMiddleware(next http.Handler) http.Handler {
// timeoutWriter wraps ResponseWriter to prevent concurrent writes after timeout // timeoutWriter wraps ResponseWriter to prevent concurrent writes after timeout
type timeoutWriter struct { type timeoutWriter struct {
http.ResponseWriter http.ResponseWriter
mu sync.Mutex mu sync.Mutex
written bool written bool
header http.Header // cached header to prevent concurrent access header http.Header // cached header to prevent concurrent access
@@ -158,6 +162,7 @@ func (tw *timeoutWriter) Header() http.Header {
func (tw *timeoutWriter) markWritten() { func (tw *timeoutWriter) markWritten() {
tw.mu.Lock() tw.mu.Lock()
defer tw.mu.Unlock() defer tw.mu.Unlock()
tw.written = true tw.written = true
} }
@@ -168,7 +173,10 @@ func (tw *timeoutWriter) markWritten() {
// The timeout parameter specifies the maximum duration allowed for request processing. // The timeout parameter specifies the maximum duration allowed for request processing.
// The returned middleware handles panics from the wrapped handler by re-panicking // The returned middleware handles panics from the wrapped handler by re-panicking
// after cleanup, and prevents concurrent writes to the response after timeout occurs. // after cleanup, and prevents concurrent writes to the response after timeout occurs.
func TimeoutMiddleware(timeout time.Duration, logger *slog.Logger) func(http.Handler) http.Handler { func TimeoutMiddleware(
timeout time.Duration,
logger *slog.Logger,
) func(http.Handler) http.Handler {
return func(next http.Handler) http.Handler { return func(next http.Handler) http.Handler {
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
startTime := time.Now() startTime := time.Now()
@@ -182,7 +190,7 @@ func TimeoutMiddleware(timeout time.Duration, logger *slog.Logger) func(http.Han
} }
done := make(chan struct{}) done := make(chan struct{})
panicChan := make(chan interface{}, 1) panicChan := make(chan any, 1)
go func() { go func() {
defer func() { defer func() {
@@ -202,6 +210,7 @@ func TimeoutMiddleware(timeout time.Duration, logger *slog.Logger) func(http.Han
return return
case <-ctx.Done(): case <-ctx.Done():
tw.markWritten() // Prevent the handler from writing after timeout tw.markWritten() // Prevent the handler from writing after timeout
execTime := time.Since(startTime) execTime := time.Since(startTime)
// Log the timeout as a warning // Log the timeout as a warning
@@ -219,13 +228,13 @@ func TimeoutMiddleware(timeout time.Duration, logger *slog.Logger) func(http.Han
w.WriteHeader(http.StatusRequestTimeout) w.WriteHeader(http.StatusRequestTimeout)
tw.mu.Unlock() tw.mu.Unlock()
response := map[string]interface{}{ response := map[string]any{
"status": "error", jsonKeyStatus: jsonStatusError,
"error": map[string]interface{}{ jsonKeyError: map[string]any{
"msg": "Request timeout", jsonKeyMsg: "Request timeout",
"code": http.StatusRequestTimeout, jsonKeyCode: http.StatusRequestTimeout,
}, },
"@meta": map[string]interface{}{ "@meta": map[string]any{
"time_zone": "UTC", "time_zone": "UTC",
"api_version": 1, "api_version": 1,
"execution_time": fmt.Sprintf("%d ms", execTime.Milliseconds()), "execution_time": fmt.Sprintf("%d ms", execTime.Milliseconds()),
@@ -273,8 +282,10 @@ func JSONValidationMiddleware(next http.Handler) http.Handler {
// For JSON responses, validate the JSON // For JSON responses, validate the JSON
if rw.body.Len() > 0 { if rw.body.Len() > 0 {
var testParse interface{} var testParse any
if err := json.Unmarshal(rw.body.Bytes(), &testParse); err == nil {
err := json.Unmarshal(rw.body.Bytes(), &testParse)
if err == nil {
// Valid JSON, write it out // Valid JSON, write it out
w.Header().Set("Content-Type", "application/json") w.Header().Set("Content-Type", "application/json")
w.WriteHeader(rw.statusCode) w.WriteHeader(rw.statusCode)
@@ -293,6 +304,7 @@ func JSONValidationMiddleware(next http.Handler) http.Handler {
if statusCode == http.StatusOK { if statusCode == http.StatusOK {
statusCode = http.StatusInternalServerError statusCode = http.StatusInternalServerError
} }
w.WriteHeader(statusCode) w.WriteHeader(statusCode)
errorMsg := "Internal server error" errorMsg := "Internal server error"
@@ -300,11 +312,11 @@ func JSONValidationMiddleware(next http.Handler) http.Handler {
errorMsg = "Request timeout" errorMsg = "Request timeout"
} }
response := map[string]interface{}{ response := map[string]any{
"status": "error", jsonKeyStatus: jsonStatusError,
"error": map[string]interface{}{ jsonKeyError: map[string]any{
"msg": errorMsg, jsonKeyMsg: errorMsg,
"code": statusCode, jsonKeyCode: statusCode,
}, },
} }
_ = json.NewEncoder(w).Encode(response) _ = json.NewEncoder(w).Encode(response)
@@ -314,6 +326,7 @@ func JSONValidationMiddleware(next http.Handler) http.Handler {
// statusWriter wraps http.ResponseWriter to capture the status code // statusWriter wraps http.ResponseWriter to capture the status code
type statusWriter struct { type statusWriter struct {
http.ResponseWriter http.ResponseWriter
statusCode int statusCode int
written bool written bool
} }
@@ -323,6 +336,7 @@ func (sw *statusWriter) WriteHeader(statusCode int) {
sw.statusCode = statusCode sw.statusCode = statusCode
sw.written = true sw.written = true
} }
sw.ResponseWriter.WriteHeader(statusCode) sw.ResponseWriter.WriteHeader(statusCode)
} }
@@ -362,11 +376,12 @@ func RequestLoggerMiddleware(logger *slog.Logger) func(http.Handler) http.Handle
// Slow query threshold (1 second) // Slow query threshold (1 second)
const slowQueryThreshold = 1 * time.Second const slowQueryThreshold = 1 * time.Second
if sw.statusCode >= http.StatusInternalServerError { switch {
case sw.statusCode >= http.StatusInternalServerError:
logLevel = slog.LevelError logLevel = slog.LevelError
} else if sw.statusCode >= http.StatusBadRequest { case sw.statusCode >= http.StatusBadRequest:
logLevel = slog.LevelWarn logLevel = slog.LevelWarn
} else if duration >= slowQueryThreshold { case duration >= slowQueryThreshold:
logLevel = slog.LevelWarn logLevel = slog.LevelWarn
} }
internal/server/routes.go
+6 -1
View File
@@ -16,7 +16,9 @@ func (s *Server) setupRoutes() {
r.Use(middleware.RealIP) r.Use(middleware.RealIP)
r.Use(RequestLoggerMiddleware(s.logger.Logger)) // Structured request logging r.Use(RequestLoggerMiddleware(s.logger.Logger)) // Structured request logging
r.Use(middleware.Recoverer) r.Use(middleware.Recoverer)
const requestTimeout = 30 * time.Second // Increased from 8s for slow queries const requestTimeout = 30 * time.Second // Increased from 8s for slow queries
r.Use(TimeoutMiddleware(requestTimeout, s.logger.Logger)) r.Use(TimeoutMiddleware(requestTimeout, s.logger.Logger))
r.Use(JSONResponseMiddleware) r.Use(JSONResponseMiddleware)
@@ -24,7 +26,10 @@ func (s *Server) setupRoutes() {
r.Get("/", s.handleIndex()) r.Get("/", s.handleIndex())
r.Get("/status", s.handleStatusHTML()) r.Get("/status", s.handleStatusHTML())
r.Get("/status.json", JSONValidationMiddleware(s.handleStatusJSON()).ServeHTTP) r.Get("/status.json", JSONValidationMiddleware(s.handleStatusJSON()).ServeHTTP)
r.Get("/.well-known/healthcheck.json", JSONValidationMiddleware(s.handleHealthCheck()).ServeHTTP) r.Get(
"/.well-known/healthcheck.json",
JSONValidationMiddleware(s.handleHealthCheck()).ServeHTTP,
)
// AS and prefix detail pages // AS and prefix detail pages
r.Get("/as/{asn}", s.handleASDetail()) r.Get("/as/{asn}", s.handleASDetail())
internal/server/server.go
+31 -14
View File
@@ -3,10 +3,14 @@ package server
import ( import (
"context" "context"
"errors"
"fmt"
"net"
"net/http" "net/http"
"os" "strconv"
"time" "time"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/database" "git.eeqj.de/sneak/routewatch/internal/database"
"git.eeqj.de/sneak/routewatch/internal/logger" "git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/streamer" "git.eeqj.de/sneak/routewatch/internal/streamer"
@@ -33,16 +37,23 @@ type Server struct {
db database.Store db database.Store
streamer *streamer.Streamer streamer *streamer.Streamer
logger *logger.Logger logger *logger.Logger
port int
srv *http.Server srv *http.Server
asnFetcher ASNFetcher asnFetcher ASNFetcher
} }
// New creates a new HTTP server // New creates a new HTTP server
func New(db database.Store, streamer *streamer.Streamer, logger *logger.Logger) *Server { func New(
db database.Store,
streamer *streamer.Streamer,
logger *logger.Logger,
cfg *config.Config,
) *Server {
s := &Server{ s := &Server{
db: db, db: db,
streamer: streamer, streamer: streamer,
logger: logger, logger: logger,
port: cfg.Port,
} }
s.setupRoutes() s.setupRoutes()
@@ -50,14 +61,12 @@ func New(db database.Store, streamer *streamer.Streamer, logger *logger.Logger)
return s return s
} }
// Start starts the HTTP server // Start listens on the configured port and serves HTTP in the background.
func (s *Server) Start() error { // It returns an error if it cannot listen, for example because the port is
port := os.Getenv("PORT") // already in use. ctx bounds only opening the port, not serving.
if port == "" { func (s *Server) Start(ctx context.Context) error {
port = "8080"
}
const ( const (
network = "tcp"
readHeaderTimeout = 40 * time.Second readHeaderTimeout = 40 * time.Second
readTimeout = 60 * time.Second readTimeout = 60 * time.Second
writeTimeout = 60 * time.Second writeTimeout = 60 * time.Second
@@ -65,7 +74,7 @@ func (s *Server) Start() error {
) )
s.srv = &http.Server{ s.srv = &http.Server{
Addr: ":" + port, Addr: ":" + strconv.Itoa(s.port),
Handler: s.router, Handler: s.router,
ReadHeaderTimeout: readHeaderTimeout, ReadHeaderTimeout: readHeaderTimeout,
ReadTimeout: readTimeout, ReadTimeout: readTimeout,
@@ -73,12 +82,20 @@ func (s *Server) Start() error {
IdleTimeout: idleTimeout, IdleTimeout: idleTimeout,
} }
s.logger.Info("Starting HTTP server", "port", port, "addr", s.srv.Addr) s.logger.Info("Starting HTTP server", "port", s.port, "addr", s.srv.Addr)
var listenConfig net.ListenConfig
listener, err := listenConfig.Listen(ctx, network, s.srv.Addr)
if err != nil {
return fmt.Errorf("starting HTTP server: %w", err)
}
s.logger.Info("HTTP server listening", "addr", listener.Addr().String())
// Start in goroutine but log when actually listening
go func() { go func() {
s.logger.Info("HTTP server listening", "addr", s.srv.Addr) err := s.srv.Serve(listener)
if err := s.srv.ListenAndServe(); err != nil && err != http.ErrServerClosed { if err != nil && !errors.Is(err, http.ErrServerClosed) {
s.logger.Error("HTTP server error", "error", err) s.logger.Error("HTTP server error", "error", err)
} }
}() }()
internal/server/server_test.go
+47
View File
@@ -0,0 +1,47 @@
package server_test
import (
"context"
"net"
"testing"
"git.eeqj.de/sneak/routewatch/internal/config"
"git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/metrics"
"git.eeqj.de/sneak/routewatch/internal/server"
"git.eeqj.de/sneak/routewatch/internal/streamer"
)
// TestStartFailsWhenPortIsHeld holds a port with a listener of its own and
// checks that Start on that port returns an error. Before the fix Start
// listened in a goroutine, only logged the failure and returned nil, so the
// daemon kept running without its web server.
func TestStartFailsWhenPortIsHeld(t *testing.T) {
t.Parallel()
var listenConfig net.ListenConfig
held, err := listenConfig.Listen(t.Context(), "tcp", ":0")
if err != nil {
t.Fatalf("failed to open the test listener: %v", err)
}
defer func() { _ = held.Close() }()
addr, ok := held.Addr().(*net.TCPAddr)
if !ok {
t.Fatalf("test listener address %v is not a TCP address", held.Addr())
}
port := addr.Port
log := logger.New()
s := server.New(nil, streamer.New(log, metrics.New()), log,
&config.Config{Port: port})
defer func() { _ = s.Stop(context.Background()) }()
err = s.Start(t.Context())
if err == nil {
t.Fatalf("Start returned nil while port %d was held by another listener",
port)
}
}
internal/streamer/streamer.go
+329 -219
View File
@@ -1,5 +1,6 @@
// Package streamer implements an HTTP client that connects to the RIPE RIS Live streaming API, // Package streamer implements an HTTP client that connects to the RIPE RIS Live
// parses BGP UPDATE messages from the JSON stream, and dispatches them to registered handlers. // streaming API, parses BGP UPDATE messages from the JSON stream, and
// dispatches them to registered handlers.
package streamer package streamer
import ( import (
@@ -7,6 +8,7 @@ import (
"compress/gzip" "compress/gzip"
"context" "context"
"encoding/json" "encoding/json"
"errors"
"fmt" "fmt"
"io" "io"
"math" "math"
@@ -24,30 +26,32 @@ import (
// countingReader wraps an io.Reader and counts bytes read // countingReader wraps an io.Reader and counts bytes read
type countingReader struct { type countingReader struct {
reader io.Reader reader io.Reader
count int64 count atomic.Int64
} }
// Read implements io.Reader and counts bytes // Read implements io.Reader and counts bytes
func (c *countingReader) Read(p []byte) (int, error) { func (c *countingReader) Read(p []byte) (int, error) {
n, err := c.reader.Read(p) n, err := c.reader.Read(p)
atomic.AddInt64(&c.count, int64(n)) c.count.Add(int64(n))
return n, err return n, err
} }
// Count returns the total bytes read // Count returns the total bytes read
func (c *countingReader) Count() int64 { func (c *countingReader) Count() int64 {
return atomic.LoadInt64(&c.count) return c.count.Load()
} }
// Configuration constants for the RIS Live streamer. // Configuration constants for the RIS Live streamer.
const ( const (
risLiveURL = "https://ris-live.ripe.net/v1/stream/?format=json&" + risLiveURL = "https://ris-live.ripe.net/v1/stream/?format=json&" +
"client=https%3A%2F%2Fgit.eeqj.de%2Fsneak%2Froutewatch" "client=https%3A%2F%2Fgit.eeqj.de%2Fsneak%2Froutewatch"
metricsWindowSize = 60 // seconds for rolling average metricsWindowSize = 60 // seconds for rolling average
metricsUpdateRate = time.Second metricsUpdateRate = time.Second
minBackoffDelay = 5 * time.Second minBackoffDelay = 5 * time.Second
maxBackoffDelay = 320 * time.Second maxBackoffDelay = 320 * time.Second
// Reconnect when no line arrives for this long
streamIdleTimeout = 60 * time.Second
metricsLogInterval = 10 * time.Second metricsLogInterval = 10 * time.Second
bytesPerKB = 1024 bytesPerKB = 1024
bytesPerMB = 1024 * 1024 bytesPerMB = 1024 * 1024
@@ -58,11 +62,22 @@ const (
backpressureSlope = 2.0 // Slope for linear drop probability increase backpressureSlope = 2.0 // Slope for linear drop probability increase
) )
var (
errAlreadyRunning = errors.New("streamer already running")
errUnexpectedStatus = errors.New("unexpected status code")
// errStreamEnded is returned by stream when RIS Live ends the response.
errStreamEnded = errors.New("RIS Live ended the stream")
// errStreamIdle is returned by stream when no line arrived for the idle
// timeout.
errStreamIdle = errors.New("no line received")
)
// MessageHandler defines the interface for processing RIS messages. // MessageHandler defines the interface for processing RIS messages.
// Implementations must specify which message types they want to receive, // Implementations must specify which message types they want to receive,
// how to process messages, and their desired queue capacity. // how to process messages, and their desired queue capacity.
type MessageHandler interface { type MessageHandler interface {
// WantsMessage returns true if this handler wants to process messages of the given type. // WantsMessage returns true if this handler wants to process messages of
// the given type.
WantsMessage(messageType string) bool WantsMessage(messageType string) bool
// HandleMessage processes a RIS message. This method is called from a dedicated // HandleMessage processes a RIS message. This method is called from a dedicated
@@ -104,16 +119,20 @@ type handlerInfo struct {
// backoff, dispatches messages to registered handlers via per-handler queues, and // backoff, dispatches messages to registered handlers via per-handler queues, and
// implements backpressure to prevent queue overflow during high traffic periods. // implements backpressure to prevent queue overflow during high traffic periods.
type Streamer struct { type Streamer struct {
logger *logger.Logger logger *logger.Logger
client *http.Client client *http.Client
url string url string
// Drop the connection when no line arrives for this long
idleTimeout time.Duration
// First wait before reconnecting; doubles up to maxBackoffDelay
minBackoff time.Duration
handlers []*handlerInfo handlers []*handlerInfo
rawHandler RawMessageHandler rawHandler RawMessageHandler
mu sync.RWMutex mu sync.RWMutex
cancel context.CancelFunc cancel context.CancelFunc
running bool running bool
metrics *metrics.Tracker metrics *metrics.Tracker
totalDropped uint64 // Total dropped messages across all handlers totalDropped atomic.Uint64 // Total dropped messages across all handlers
random *rand.Rand // Random number generator for backpressure drops random *rand.Rand // Random number generator for backpressure drops
bgpPeers map[string]bool // Track active BGP peers by peer IP bgpPeers map[string]bool // Track active BGP peers by peer IP
bgpPeersMu sync.RWMutex // Protects bgpPeers map bgpPeersMu sync.RWMutex // Protects bgpPeers map
@@ -121,11 +140,14 @@ type Streamer struct {
// New creates a new Streamer instance configured to connect to the RIS Live API. // New creates a new Streamer instance configured to connect to the RIS Live API.
// The logger is used for structured logging of connection events and errors. // The logger is used for structured logging of connection events and errors.
// The metrics tracker is used to record message counts, bytes received, and connection status. // The metrics tracker is used to record message counts, bytes received, and
// connection status.
func New(logger *logger.Logger, metrics *metrics.Tracker) *Streamer { func New(logger *logger.Logger, metrics *metrics.Tracker) *Streamer {
return &Streamer{ return &Streamer{
logger: logger, logger: logger,
url: risLiveURL, url: risLiveURL,
idleTimeout: streamIdleTimeout,
minBackoff: minBackoffDelay,
client: &http.Client{ client: &http.Client{
Timeout: 0, // No timeout for streaming Timeout: 0, // No timeout for streaming
Transport: &http.Transport{ Transport: &http.Transport{
@@ -153,7 +175,8 @@ func (s *Streamer) RegisterHandler(handler MessageHandler) {
handler: handler, handler: handler,
queue: make(chan *ristypes.RISMessage, handler.QueueCapacity()), queue: make(chan *ristypes.RISMessage, handler.QueueCapacity()),
metrics: handlerMetrics{ metrics: handlerMetrics{
minTime: time.Duration(math.MaxInt64), // Initialize to max so first value sets the floor // Initialize to max so first value sets the floor
minTime: time.Duration(math.MaxInt64),
}, },
} }
@@ -171,22 +194,24 @@ func (s *Streamer) RegisterHandler(handler MessageHandler) {
func (s *Streamer) RegisterRawHandler(handler RawMessageHandler) { func (s *Streamer) RegisterRawHandler(handler RawMessageHandler) {
s.mu.Lock() s.mu.Lock()
defer s.mu.Unlock() defer s.mu.Unlock()
s.rawHandler = handler s.rawHandler = handler
} }
// Start begins streaming BGP updates from the RIS Live API in a background goroutine. // Start begins streaming BGP updates from the RIS Live API in a background goroutine.
// It starts worker goroutines for each registered handler and manages automatic // It starts worker goroutines for each registered handler and manages automatic
// reconnection with exponential backoff on connection failures. // reconnection with exponential backoff on connection failures.
// Returns an error if the streamer is already running. // Returns an error if the streamer is already running. It streams until Stop:
func (s *Streamer) Start() error { // the end of ctx does not stop it.
func (s *Streamer) Start(ctx context.Context) error {
s.mu.Lock() s.mu.Lock()
defer s.mu.Unlock() defer s.mu.Unlock()
if s.running { if s.running {
return fmt.Errorf("streamer already running") return errAlreadyRunning
} }
ctx, cancel := context.WithCancel(context.Background()) ctx, cancel := context.WithCancel(context.WithoutCancel(ctx))
s.cancel = cancel s.cancel = cancel
s.running = true s.running = true
@@ -210,45 +235,27 @@ func (s *Streamer) Start() error {
// the connection status in metrics. This method is safe to call multiple times. // the connection status in metrics. This method is safe to call multiple times.
func (s *Streamer) Stop() { func (s *Streamer) Stop() {
s.mu.Lock() s.mu.Lock()
if s.cancel != nil { if s.cancel == nil {
s.cancel() // Not started, or already stopped: closing the queues again would panic.
s.mu.Unlock()
return
} }
s.cancel()
s.cancel = nil
// Close all handler queues to signal workers to stop // Close all handler queues to signal workers to stop
for _, info := range s.handlers { for _, info := range s.handlers {
close(info.queue) close(info.queue)
} }
s.running = false s.running = false
s.mu.Unlock() s.mu.Unlock()
s.metrics.SetConnected(false) s.metrics.SetConnected(false)
} }
// runHandlerWorker processes messages for a specific handler // IsRunning reports whether the streamer is currently connected and processing
func (s *Streamer) runHandlerWorker(info *handlerInfo) { // messages. This is safe to call concurrently from multiple goroutines.
for msg := range info.queue {
start := time.Now()
info.handler.HandleMessage(msg)
elapsed := time.Since(start)
// Update metrics
info.metrics.mu.Lock()
info.metrics.processedCount++
info.metrics.totalTime += elapsed
// Update min time
if elapsed < info.metrics.minTime {
info.metrics.minTime = elapsed
}
// Update max time
if elapsed > info.metrics.maxTime {
info.metrics.maxTime = elapsed
}
info.metrics.mu.Unlock()
}
}
// IsRunning reports whether the streamer is currently connected and processing messages.
// This is safe to call concurrently from multiple goroutines.
func (s *Streamer) IsRunning() bool { func (s *Streamer) IsRunning() bool {
s.mu.RLock() s.mu.RLock()
defer s.mu.RUnlock() defer s.mu.RUnlock()
@@ -309,12 +316,15 @@ func (s *Streamer) GetHandlerStats() []HandlerStats {
// Calculate average time // Calculate average time
if info.metrics.processedCount > 0 { if info.metrics.processedCount > 0 {
processedCount := info.metrics.processedCount processedCount := info.metrics.processedCount
const maxInt64 = 1<<63 - 1 const maxInt64 = 1<<63 - 1
if processedCount > maxInt64 { if processedCount > maxInt64 {
processedCount = maxInt64 processedCount = maxInt64
} }
//nolint:gosec // processedCount is explicitly bounded above
hs.AvgProcessTime = info.metrics.totalTime / time.Duration(processedCount) hs.AvgProcessTime = info.metrics.totalTime /
time.Duration(processedCount)
} }
info.metrics.mu.Unlock() info.metrics.mu.Unlock()
@@ -325,11 +335,38 @@ func (s *Streamer) GetHandlerStats() []HandlerStats {
return stats return stats
} }
// GetDroppedMessages returns the total number of messages dropped across all handlers // GetDroppedMessages returns the total number of messages dropped across all
// due to queue overflow or backpressure. This counter is monotonically increasing // handlers due to queue overflow or backpressure. This counter is monotonically
// and is safe to call concurrently. // increasing and is safe to call concurrently.
func (s *Streamer) GetDroppedMessages() uint64 { func (s *Streamer) GetDroppedMessages() uint64 {
return atomic.LoadUint64(&s.totalDropped) return s.totalDropped.Load()
}
// runHandlerWorker processes messages for a specific handler
func (s *Streamer) runHandlerWorker(info *handlerInfo) {
for msg := range info.queue {
start := time.Now()
info.handler.HandleMessage(msg)
elapsed := time.Since(start)
// Update metrics
info.metrics.mu.Lock()
info.metrics.processedCount++
info.metrics.totalTime += elapsed
// Update min time
if elapsed < info.metrics.minTime {
info.metrics.minTime = elapsed
}
// Update max time
if elapsed > info.metrics.maxTime {
info.metrics.maxTime = elapsed
}
info.metrics.mu.Unlock()
}
} }
// logMetrics logs the current streaming statistics // logMetrics logs the current streaming statistics
@@ -338,7 +375,8 @@ func (s *Streamer) logMetrics() {
uptime := time.Since(metrics.ConnectedSince) uptime := time.Since(metrics.ConnectedSince)
const bitsPerMegabit = 1000000 const bitsPerMegabit = 1000000
totalDropped := atomic.LoadUint64(&s.totalDropped)
totalDropped := s.totalDropped.Load()
s.logger.Info( s.logger.Info(
"Stream statistics", "Stream statistics",
@@ -364,16 +402,19 @@ func (s *Streamer) logMetrics() {
// Log per-handler statistics // Log per-handler statistics
s.mu.RLock() s.mu.RLock()
for i, info := range s.handlers { for i, info := range s.handlers {
info.metrics.mu.Lock() info.metrics.mu.Lock()
if info.metrics.processedCount > 0 { if info.metrics.processedCount > 0 {
// Safe conversion: processedCount is bounded by maxInt64 // Safe conversion: processedCount is bounded by maxInt64
processedCount := info.metrics.processedCount processedCount := info.metrics.processedCount
const maxInt64 = 1<<63 - 1 const maxInt64 = 1<<63 - 1
if processedCount > maxInt64 { if processedCount > maxInt64 {
processedCount = maxInt64 processedCount = maxInt64
} }
//nolint:gosec // processedCount is explicitly bounded above
avgTime := info.metrics.totalTime / time.Duration(processedCount) avgTime := info.metrics.totalTime / time.Duration(processedCount)
s.logger.Info( s.logger.Info(
"Handler statistics", "Handler statistics",
@@ -390,6 +431,7 @@ func (s *Streamer) logMetrics() {
} }
info.metrics.mu.Unlock() info.metrics.mu.Unlock()
} }
s.mu.RUnlock() s.mu.RUnlock()
} }
@@ -398,9 +440,10 @@ func (s *Streamer) updateMetrics(messageBytes int) {
s.metrics.RecordMessage(int64(messageBytes)) s.metrics.RecordMessage(int64(messageBytes))
} }
// streamWithReconnect handles streaming with automatic reconnection and exponential backoff // streamWithReconnect handles streaming with automatic reconnection and
// exponential backoff
func (s *Streamer) streamWithReconnect(ctx context.Context) { func (s *Streamer) streamWithReconnect(ctx context.Context) {
backoffDelay := minBackoffDelay backoffDelay := s.minBackoff
consecutiveFailures := 0 consecutiveFailures := 0
for { for {
@@ -417,8 +460,9 @@ func (s *Streamer) streamWithReconnect(ctx context.Context) {
err := s.stream(ctx) err := s.stream(ctx)
streamDuration := time.Since(startTime) streamDuration := time.Since(startTime)
if err == nil { // Only Stop ends the streamer. A stream that ends any other way,
// Clean exit (context cancelled) // with or without an error, is a disconnect and is reconnected.
if ctx.Err() != nil {
return return
} }
@@ -429,17 +473,14 @@ func (s *Streamer) streamWithReconnect(ctx context.Context) {
"stream_duration", streamDuration) "stream_duration", streamDuration)
s.metrics.SetConnected(false) s.metrics.SetConnected(false)
// Check if context is cancelled // A stream that lasted more than 30 seconds resets the backoff. That
if ctx.Err() != nil { // includes a stream dropped after streamIdleTimeout with no line, so
return // a silent feed is retried after the first backoff delay every time.
}
// If we streamed for more than 30 seconds, reset the backoff
// This indicates we had a successful connection that received data
if streamDuration > 30*time.Second { if streamDuration > 30*time.Second {
s.logger.Info("Resetting backoff delay due to successful connection", s.logger.Info("Resetting backoff delay due to successful connection",
"stream_duration", streamDuration) "stream_duration", streamDuration)
backoffDelay = minBackoffDelay
backoffDelay = s.minBackoff
consecutiveFailures = 0 consecutiveFailures = 0
} else { } else {
// Increment consecutive failures // Increment consecutive failures
@@ -472,7 +513,13 @@ func (s *Streamer) stream(ctx context.Context) error {
connCtx, connCancel := context.WithCancel(ctx) connCtx, connCancel := context.WithCancel(ctx)
defer connCancel() defer connCancel()
req, err := http.NewRequestWithContext(ctx, "GET", s.url, nil) // The idle timer drops the connection, by cancelling connCtx, when no
// line arrives for s.idleTimeout. Each line read starts it over. It
// also covers the wait for the server to answer.
idleTimer := time.AfterFunc(s.idleTimeout, connCancel)
defer idleTimer.Stop()
req, err := http.NewRequestWithContext(connCtx, http.MethodGet, s.url, nil)
if err != nil { if err != nil {
return fmt.Errorf("failed to create request: %w", err) return fmt.Errorf("failed to create request: %w", err)
} }
@@ -485,13 +532,14 @@ func (s *Streamer) stream(ctx context.Context) error {
return fmt.Errorf("failed to connect to RIS Live: %w", err) return fmt.Errorf("failed to connect to RIS Live: %w", err)
} }
defer func() { defer func() {
if err := resp.Body.Close(); err != nil { err := resp.Body.Close()
if err != nil {
s.logger.Error("Failed to close response body", "error", err) s.logger.Error("Failed to close response body", "error", err)
} }
}() }()
if resp.StatusCode != http.StatusOK { if resp.StatusCode != http.StatusOK {
return fmt.Errorf("unexpected status code: %d", resp.StatusCode) return fmt.Errorf("%w: %d", errUnexpectedStatus, resp.StatusCode)
} }
// Wrap body with counting reader to track actual wire bytes // Wrap body with counting reader to track actual wire bytes
@@ -505,7 +553,9 @@ func (s *Streamer) stream(ctx context.Context) error {
return fmt.Errorf("failed to create gzip reader: %w", err) return fmt.Errorf("failed to create gzip reader: %w", err)
} }
defer func() { _ = gzReader.Close() }() defer func() { _ = gzReader.Close() }()
reader = gzReader reader = gzReader
s.logger.Info("Connected to RIS Live stream", "compression", "gzip") s.logger.Info("Connected to RIS Live stream", "compression", "gzip")
} else { } else {
s.logger.Info("Connected to RIS Live stream", "compression", "none") s.logger.Info("Connected to RIS Live stream", "compression", "none")
@@ -513,9 +563,6 @@ func (s *Streamer) stream(ctx context.Context) error {
s.metrics.SetConnected(true) s.metrics.SetConnected(true)
// Track wire bytes for metrics updates
var lastWireBytes int64
// Start metrics logging goroutine // Start metrics logging goroutine
metricsTicker := time.NewTicker(metricsLogInterval) metricsTicker := time.NewTicker(metricsLogInterval)
defer metricsTicker.Stop() defer metricsTicker.Stop()
@@ -532,28 +579,49 @@ func (s *Streamer) stream(ctx context.Context) error {
}() }()
// Wire byte update ticker - update metrics with actual wire bytes periodically // Wire byte update ticker - update metrics with actual wire bytes periodically
go s.recordWireBytes(connCtx, wireCounter)
return s.readStream(ctx, connCtx, reader, idleTimer)
}
// recordWireBytes adds the bytes wireCounter has counted to the metrics once a
// second until ctx is cancelled.
func (s *Streamer) recordWireBytes(
ctx context.Context, wireCounter *countingReader,
) {
wireUpdateTicker := time.NewTicker(time.Second) wireUpdateTicker := time.NewTicker(time.Second)
defer wireUpdateTicker.Stop() defer wireUpdateTicker.Stop()
go func() { var lastWireBytes int64
for {
select {
case <-wireUpdateTicker.C:
currentBytes := wireCounter.Count()
delta := currentBytes - lastWireBytes
if delta > 0 {
s.metrics.RecordWireBytes(delta)
lastWireBytes = currentBytes
}
case <-connCtx.Done():
return
}
}
}()
for {
select {
case <-wireUpdateTicker.C:
currentBytes := wireCounter.Count()
delta := currentBytes - lastWireBytes
if delta > 0 {
s.metrics.RecordWireBytes(delta)
lastWireBytes = currentBytes
}
case <-ctx.Done():
return
}
}
}
// readStream hands each line of the stream to processLine until the stream
// ends, a line fails, or ctx is cancelled. Each line restarts idleTimer, which
// cancels connCtx when it fires.
func (s *Streamer) readStream(
ctx, connCtx context.Context, reader io.Reader, idleTimer *time.Timer,
) error {
scanner := bufio.NewScanner(reader) scanner := bufio.NewScanner(reader)
for scanner.Scan() { for scanner.Scan() {
idleTimer.Reset(s.idleTimeout)
select { select {
case <-ctx.Done(): case <-ctx.Done():
s.logger.Info("Stream stopped by context") s.logger.Info("Stream stopped by context")
@@ -567,142 +635,184 @@ func (s *Streamer) stream(ctx context.Context) error {
continue continue
} }
// Update metrics with decompressed message size err := s.processLine(ctx, line)
s.updateMetrics(len(line)) if err != nil {
return err
// Call raw handler if registered
s.mu.RLock()
rawHandler := s.rawHandler
s.mu.RUnlock()
if rawHandler != nil {
// Call raw handler synchronously to preserve order
rawHandler(string(line))
} }
// Parse the message first
var wrapper ristypes.RISLiveMessage
if err := json.Unmarshal(line, &wrapper); err != nil {
// Log the error and return to trigger reconnection
s.logger.Error("Failed to parse JSON",
"error", err,
"line", string(line),
"line_length", len(line))
return fmt.Errorf("JSON parse error: %w", err)
}
// Check if it's a ris_message wrapper
if wrapper.Type != "ris_message" {
s.logger.Error("Unexpected wrapper type",
"type", wrapper.Type,
"line", string(line),
)
continue
}
// Get the actual message
msg := wrapper.Data
// Parse the timestamp
msg.ParsedTimestamp = time.Unix(int64(msg.Timestamp), 0).UTC()
// Process based on message type
switch msg.Type {
case "UPDATE":
// Process BGP UPDATE messages
// Will be dispatched to handlers
case "RIS_PEER_STATE":
// RIS peer state messages - silently ignore
continue
case "KEEPALIVE":
// BGP keepalive messages - silently process
continue
case "OPEN":
// BGP open messages - track peer as active
s.bgpPeersMu.Lock()
s.bgpPeers[msg.Peer] = true
peerCount := len(s.bgpPeers)
s.bgpPeersMu.Unlock()
s.metrics.SetBGPPeerCount(peerCount)
s.logger.Info("BGP session opened",
"peer", msg.Peer,
"peer_asn", msg.PeerASN,
"total_peers", peerCount,
)
continue
case "NOTIFICATION":
// BGP notification messages (session closed)
s.bgpPeersMu.Lock()
delete(s.bgpPeers, msg.Peer)
peerCount := len(s.bgpPeers)
s.bgpPeersMu.Unlock()
s.metrics.SetBGPPeerCount(peerCount)
s.logger.Warn("BGP notification",
"peer", msg.Peer,
"peer_asn", msg.PeerASN,
"total_peers", peerCount,
)
continue
case "STATE":
// Peer state changes - silently ignore
continue
default:
s.logger.Warn("Unknown message type, skipping",
"type", msg.Type,
)
continue
}
// Dispatch to interested handlers
s.mu.RLock()
for _, info := range s.handlers {
if !info.handler.WantsMessage(msg.Type) {
continue
}
// Check if we should drop due to backpressure
if s.shouldDropForBackpressure(info) {
atomic.AddUint64(&info.metrics.droppedCount, 1)
atomic.AddUint64(&s.totalDropped, 1)
continue
}
// Try to queue the message
select {
case info.queue <- &msg:
// Message queued successfully
// Update high water mark if needed
queueLen := len(info.queue)
info.metrics.mu.Lock()
if queueLen > info.metrics.queueHighWaterMark {
info.metrics.queueHighWaterMark = queueLen
}
info.metrics.mu.Unlock()
default:
// Queue is full, drop the message
atomic.AddUint64(&info.metrics.droppedCount, 1)
atomic.AddUint64(&s.totalDropped, 1)
}
}
s.mu.RUnlock()
} }
if err := scanner.Err(); err != nil { if ctx.Err() != nil {
return ctx.Err()
}
if connCtx.Err() != nil {
return fmt.Errorf("%w for %s", errStreamIdle, s.idleTimeout)
}
err := scanner.Err()
if err != nil {
return fmt.Errorf("scanner error: %w", err) return fmt.Errorf("scanner error: %w", err)
} }
return errStreamEnded
}
// processLine passes one line of the stream to the raw handler, parses it,
// and dispatches an UPDATE message to the handlers. An error ends the
// connection.
func (s *Streamer) processLine(ctx context.Context, line []byte) error {
// Update metrics with decompressed message size
s.updateMetrics(len(line))
// Call raw handler if registered
s.mu.RLock()
rawHandler := s.rawHandler
s.mu.RUnlock()
if rawHandler != nil {
// Call raw handler synchronously to preserve order
rawHandler(string(line))
}
// Parse the message first
var wrapper ristypes.RISLiveMessage
err := json.Unmarshal(line, &wrapper)
if err != nil {
// Log the error and return to trigger reconnection
s.logger.Error("Failed to parse JSON",
"error", err,
"line", string(line),
"line_length", len(line))
return fmt.Errorf("JSON parse error: %w", err)
}
// Check if it's a ris_message wrapper
if wrapper.Type != "ris_message" {
s.logger.Error("Unexpected wrapper type",
"type", wrapper.Type,
"line", string(line),
)
return nil
}
// Get the actual message
msg := wrapper.Data
// Parse the timestamp
msg.ParsedTimestamp = time.Unix(int64(msg.Timestamp), 0).UTC()
// Process based on message type
switch msg.Type {
case "UPDATE":
// Process BGP UPDATE messages
// Will be dispatched to handlers
case "RIS_PEER_STATE":
// RIS peer state messages - silently ignore
return nil
case "KEEPALIVE":
// BGP keepalive messages - silently process
return nil
case "OPEN":
// BGP open messages - track peer as active
s.bgpPeersMu.Lock()
s.bgpPeers[msg.Peer] = true
peerCount := len(s.bgpPeers)
s.bgpPeersMu.Unlock()
s.metrics.SetBGPPeerCount(peerCount)
s.logger.Info("BGP session opened",
"peer", msg.Peer,
"peer_asn", msg.PeerASN,
"total_peers", peerCount,
)
return nil
case "NOTIFICATION":
// BGP notification messages (session closed)
s.bgpPeersMu.Lock()
delete(s.bgpPeers, msg.Peer)
peerCount := len(s.bgpPeers)
s.bgpPeersMu.Unlock()
s.metrics.SetBGPPeerCount(peerCount)
s.logger.Warn("BGP notification",
"peer", msg.Peer,
"peer_asn", msg.PeerASN,
"total_peers", peerCount,
)
return nil
case "STATE":
// Peer state changes - silently ignore
return nil
default:
s.logger.Warn("Unknown message type, skipping",
"type", msg.Type,
)
return nil
}
return s.dispatchMessage(ctx, &msg)
}
// dispatchMessage queues msg for each handler that wants it, dropping it for
// a handler whose queue is full or under backpressure.
func (s *Streamer) dispatchMessage(
ctx context.Context, msg *ristypes.RISMessage,
) error {
// Stop cancels ctx and closes the queues under the write lock, so if ctx
// is cancelled here, under the read lock, the queues are closed and must
// not be sent to.
s.mu.RLock()
if ctx.Err() != nil {
s.mu.RUnlock()
return ctx.Err()
}
for _, info := range s.handlers {
if !info.handler.WantsMessage(msg.Type) {
continue
}
// Check if we should drop due to backpressure
if s.shouldDropForBackpressure(info) {
atomic.AddUint64(&info.metrics.droppedCount, 1)
s.totalDropped.Add(1)
continue
}
// Try to queue the message
select {
case info.queue <- msg:
// Message queued successfully
// Update high water mark if needed
queueLen := len(info.queue)
info.metrics.mu.Lock()
if queueLen > info.metrics.queueHighWaterMark {
info.metrics.queueHighWaterMark = queueLen
}
info.metrics.mu.Unlock()
default:
// Queue is full, drop the message
atomic.AddUint64(&info.metrics.droppedCount, 1)
s.totalDropped.Add(1)
}
}
s.mu.RUnlock()
return nil return nil
} }
// shouldDropForBackpressure determines if a message should be dropped based on queue utilization // shouldDropForBackpressure determines if a message should be dropped based on
// queue utilization
func (s *Streamer) shouldDropForBackpressure(info *handlerInfo) bool { func (s *Streamer) shouldDropForBackpressure(info *handlerInfo) bool {
// Calculate queue utilization // Calculate queue utilization
queueLen := len(info.queue) queueLen := len(info.queue)
internal/streamer/streamer_internal_test.go
+332
View File
@@ -0,0 +1,332 @@
package streamer
import (
"context"
"errors"
"io"
"net/http"
"net/http/httptest"
"runtime"
"sync/atomic"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/metrics"
"git.eeqj.de/sneak/routewatch/internal/ristypes"
)
func TestNewStreamer(t *testing.T) {
t.Parallel()
logger := logger.New()
metricsTracker := metrics.New()
s := New(logger, metricsTracker)
if s == nil {
t.Fatal("New() returned nil")
}
if s.logger != logger {
t.Error("logger not set correctly")
}
if s.client == nil {
t.Error("HTTP client not initialized")
}
if s.handlers == nil {
t.Error("handlers slice not initialized")
}
if s.metrics != metricsTracker {
t.Error("metrics tracker not set correctly")
}
if s.idleTimeout != streamIdleTimeout {
t.Errorf("idleTimeout = %s, want %s", s.idleTimeout, streamIdleTimeout)
}
if s.minBackoff != minBackoffDelay {
t.Errorf("minBackoff = %s, want %s", s.minBackoff, minBackoffDelay)
}
}
// TestStreamDoesNotLeakTickersAcrossReconnects drives many short-lived
// connections (each stream call is one reconnect cycle) and asserts the
// goroutine count returns to its starting value. Each connection starts two
// ticker goroutines; before the fix they lived until the streamer's lifetime
// context was cancelled, so every reconnect leaked two.
//
//nolint:paralleltest // counts goroutines, so no other test may run alongside
func TestStreamDoesNotLeakTickersAcrossReconnects(t *testing.T) {
// The handler returns immediately, so the response body is empty and each
// stream call ends at once, standing in for a dropped connection.
srv := httptest.NewServer(http.HandlerFunc(
func(_ http.ResponseWriter, _ *http.Request) {}))
defer srv.Close()
s := New(logger.New(), metrics.New())
s.url = srv.URL
// One warm-up connection so any persistent HTTP transport goroutine exists
// before we take the baseline.
err := s.stream(context.Background())
if !errors.Is(err, errStreamEnded) {
t.Fatalf("warm-up stream returned %v, want %v", err, errStreamEnded)
}
s.client.CloseIdleConnections()
baseline := settledGoroutineCount()
const reconnects = 20
for range reconnects {
err := s.stream(context.Background())
if !errors.Is(err, errStreamEnded) {
t.Fatalf("stream returned %v, want %v", err, errStreamEnded)
}
}
s.client.CloseIdleConnections()
if !waitForGoroutines(baseline) {
t.Fatalf("goroutines did not return to baseline %d after %d reconnects, got %d",
baseline, reconnects, runtime.NumGoroutine())
}
}
// updateLine is one UPDATE message as RIS Live sends it.
const updateLine = `{"type":"ris_message","data":{"type":"UPDATE",` +
`"peer":"192.0.2.1","peer_asn":"64496","timestamp":1700000000}}` + "\n"
// updateHandler wants UPDATE messages. For each one it handles it sends on
// handled if there is room, and otherwise does nothing; a nil handled is
// never sent on.
type updateHandler struct {
handled chan struct{}
}
func (updateHandler) WantsMessage(messageType string) bool {
return messageType == "UPDATE"
}
func (updateHandler) QueueCapacity() int { return 10 }
func (h updateHandler) HandleMessage(*ristypes.RISMessage) {
select {
case h.handled <- struct{}{}:
default:
}
}
// flush sends what w has buffered to the client. The writers httptest servers
// hand to handlers can always do that.
func flush(w http.ResponseWriter) {
flusher, ok := w.(http.Flusher)
if ok {
flusher.Flush()
}
}
// TestStopBeforeMessageReachesQueues stops the streamer after the read loop
// has checked for cancellation but before it hands the message to the handler
// queues. That is the gap a stop from another goroutine can land in, and it
// used to end in "send on closed channel". The raw handler runs in that gap on
// the read loop itself, so calling Stop from it hits the gap every time.
func TestStopBeforeMessageReachesQueues(t *testing.T) {
t.Parallel()
srv := httptest.NewServer(http.HandlerFunc(
func(w http.ResponseWriter, _ *http.Request) {
_, _ = io.WriteString(w, updateLine)
}))
defer srv.Close()
s := New(logger.New(), metrics.New())
s.url = srv.URL
s.RegisterHandler(updateHandler{})
s.RegisterRawHandler(func(string) { s.Stop() })
// Start would run the stream in the background, where the test cannot
// wait for it. Setting cancel as Start does lets Stop cancel the stream
// run here instead.
ctx, cancel := context.WithCancel(context.Background())
s.cancel = cancel
err := s.stream(ctx)
if !errors.Is(err, context.Canceled) {
t.Fatalf("stream returned %v, want %v", err, context.Canceled)
}
// A second Stop must not close the queues again.
s.Stop()
}
// TestReconnectWhenStreamEnds has the server end the first response
// normally, as when RIS Live restarts. The streamer must treat that as a
// disconnect and connect again.
func TestReconnectWhenStreamEnds(t *testing.T) {
t.Parallel()
endResponse := func(http.ResponseWriter, *http.Request) {}
checkReconnect(t, endResponse, streamIdleTimeout)
}
// TestReconnectWhenStreamGoesIdle has the server hold the first response
// open without sending anything. The streamer must drop it once the idle
// timeout passes and connect again.
func TestReconnectWhenStreamGoesIdle(t *testing.T) {
t.Parallel()
const idleTimeout = 50 * time.Millisecond
holdResponse := func(w http.ResponseWriter, r *http.Request) {
flush(w)
<-r.Context().Done()
}
checkReconnect(t, holdResponse, idleTimeout)
}
// TestReconnectWhenServerNeverAnswers has the server accept the first
// request and never answer it, not even with headers. The streamer must give
// up on it once the idle timeout passes and connect again.
func TestReconnectWhenServerNeverAnswers(t *testing.T) {
t.Parallel()
const idleTimeout = 50 * time.Millisecond
neverAnswer := func(_ http.ResponseWriter, r *http.Request) {
<-r.Context().Done()
}
checkReconnect(t, neverAnswer, idleTimeout)
}
// TestStreamNotDroppedWhileLinesArrive has the server send lines well within
// the idle timeout for several times the timeout, then end the response. Each
// line must start the idle timeout over, so the stream lasts until the server
// ends it instead of being dropped as idle.
func TestStreamNotDroppedWhileLinesArrive(t *testing.T) {
t.Parallel()
const (
idleTimeout = 100 * time.Millisecond
lineInterval = idleTimeout / 10
sendFor = 5 * idleTimeout
)
srv := httptest.NewServer(http.HandlerFunc(
func(w http.ResponseWriter, _ *http.Request) {
deadline := time.Now().Add(sendFor)
for time.Now().Before(deadline) {
_, _ = io.WriteString(w, updateLine)
flush(w)
time.Sleep(lineInterval)
}
}))
defer srv.Close()
s := New(logger.New(), metrics.New())
s.url = srv.URL
s.idleTimeout = idleTimeout
err := s.stream(context.Background())
if !errors.Is(err, errStreamEnded) {
t.Fatalf("stream returned %v, want %v", err, errStreamEnded)
}
}
// checkReconnect runs a streamer against a fake RIS Live server. The first
// response is served by firstResponse; every later one carries one UPDATE
// line and then stays open. The streamer must come back for a second
// response, must not be marked connected when it does, and must handle the
// UPDATE in it.
func checkReconnect(
t *testing.T, firstResponse http.HandlerFunc, idleTimeout time.Duration,
) {
t.Helper()
const maxWait = 5 * time.Second
tracker := metrics.New()
connectedAtReconnect := make(chan bool, 1)
var requests atomic.Int32
srv := httptest.NewServer(http.HandlerFunc(
func(w http.ResponseWriter, r *http.Request) {
if requests.Add(1) == 1 {
firstResponse(w, r)
return
}
select {
case connectedAtReconnect <- tracker.IsConnected():
default:
}
_, _ = io.WriteString(w, updateLine)
flush(w)
<-r.Context().Done()
}))
defer srv.Close()
s := New(logger.New(), tracker)
s.url = srv.URL
s.idleTimeout = idleTimeout
s.minBackoff = time.Millisecond
handler := updateHandler{handled: make(chan struct{}, 1)}
s.RegisterHandler(handler)
err := s.Start(t.Context())
if err != nil {
t.Fatalf("Start returned error: %v", err)
}
defer s.Stop()
select {
case <-handler.handled:
case <-time.After(maxWait):
t.Fatalf("no message from a second response handled within %s", maxWait)
}
if <-connectedAtReconnect {
t.Error("streamer was still marked connected when it came back " +
"for a second response")
}
}
// settledGoroutineCount lets transient goroutines finish, then reports the
// current count.
func settledGoroutineCount() int {
prev := runtime.NumGoroutine()
for range 20 {
time.Sleep(10 * time.Millisecond)
cur := runtime.NumGoroutine()
if cur == prev {
return cur
}
prev = cur
}
return prev
}
// waitForGoroutines waits until the goroutine count drops to target or below.
func waitForGoroutines(target int) bool {
for range 100 {
if runtime.NumGoroutine() <= target {
return true
}
time.Sleep(10 * time.Millisecond)
}
return false
}
internal/streamer/streamer_test.go
-104
View File
@@ -1,104 +0,0 @@
package streamer
import (
"context"
"net/http"
"net/http/httptest"
"runtime"
"testing"
"time"
"git.eeqj.de/sneak/routewatch/internal/logger"
"git.eeqj.de/sneak/routewatch/internal/metrics"
)
func TestNewStreamer(t *testing.T) {
logger := logger.New()
metricsTracker := metrics.New()
s := New(logger, metricsTracker)
if s == nil {
t.Fatal("New() returned nil")
}
if s.logger != logger {
t.Error("logger not set correctly")
}
if s.client == nil {
t.Error("HTTP client not initialized")
}
if s.handlers == nil {
t.Error("handlers slice not initialized")
}
if s.metrics != metricsTracker {
t.Error("metrics tracker not set correctly")
}
}
// TestStreamDoesNotLeakTickersAcrossReconnects drives many short-lived
// connections (each stream call is one reconnect cycle) and asserts the
// goroutine count returns to its starting value. Each connection starts two
// ticker goroutines; before the fix they lived until the streamer's lifetime
// context was cancelled, so every reconnect leaked two.
func TestStreamDoesNotLeakTickersAcrossReconnects(t *testing.T) {
// The handler returns immediately, so the response body is empty and each
// stream call ends at once, standing in for a dropped connection.
srv := httptest.NewServer(http.HandlerFunc(func(_ http.ResponseWriter, _ *http.Request) {}))
defer srv.Close()
s := New(logger.New(), metrics.New())
s.url = srv.URL
// One warm-up connection so any persistent HTTP transport goroutine exists
// before we take the baseline.
if err := s.stream(context.Background()); err != nil {
t.Fatalf("warm-up stream returned error: %v", err)
}
s.client.CloseIdleConnections()
baseline := settledGoroutineCount()
const reconnects = 20
for range reconnects {
if err := s.stream(context.Background()); err != nil {
t.Fatalf("stream returned error: %v", err)
}
}
s.client.CloseIdleConnections()
if !waitForGoroutines(baseline) {
t.Fatalf("goroutines did not return to baseline %d after %d reconnects, got %d",
baseline, reconnects, runtime.NumGoroutine())
}
}
// settledGoroutineCount lets transient goroutines finish, then reports the
// current count.
func settledGoroutineCount() int {
prev := runtime.NumGoroutine()
for range 20 {
time.Sleep(10 * time.Millisecond)
cur := runtime.NumGoroutine()
if cur == prev {
return cur
}
prev = cur
}
return prev
}
// waitForGoroutines waits until the goroutine count drops to target or below.
func waitForGoroutines(target int) bool {
for range 100 {
if runtime.NumGoroutine() <= target {
return true
}
time.Sleep(10 * time.Millisecond)
}
return false
}
internal/templates/prefix_detail.html
+1
View File
@@ -283,6 +283,7 @@
<div class="origin-item"> <div class="origin-item">
<a href="/as/{{.ASN}}" class="as-link">AS{{.ASN}}</a> <a href="/as/{{.ASN}}" class="as-link">AS{{.ASN}}</a>
{{if .Handle}} ({{.Handle}}){{end}} {{if .Handle}} ({{.Handle}}){{end}}
{{if .Description}} {{.Description}}{{end}}
<span style="color: #7f8c8d; margin-left: 10px;">{{.PeerCount}} peer{{if ne .PeerCount 1}}s{{end}}</span> <span style="color: #7f8c8d; margin-left: 10px;">{{.PeerCount}} peer{{if ne .PeerCount 1}}s{{end}}</span>
</div> </div>
{{end}} {{end}}
internal/templates/templates.go
+11 -4
View File
@@ -60,6 +60,7 @@ func timeSince(t time.Time) string {
if duration < time.Minute { if duration < time.Minute {
return "just now" return "just now"
} }
if duration < time.Hour { if duration < time.Hour {
minutes := int(duration.Minutes()) minutes := int(duration.Minutes())
if minutes == 1 { if minutes == 1 {
@@ -68,6 +69,7 @@ func timeSince(t time.Time) string {
return duration.Truncate(time.Minute).String() + " ago" return duration.Truncate(time.Minute).String() + " ago"
} }
if duration < hoursPerDay*time.Hour { if duration < hoursPerDay*time.Hour {
hours := int(duration.Hours()) hours := int(duration.Hours())
if hours == 1 { if hours == 1 {
@@ -76,10 +78,12 @@ func timeSince(t time.Time) string {
return duration.Truncate(time.Hour).String() + " ago" return duration.Truncate(time.Hour).String() + " ago"
} }
days := int(duration.Hours() / hoursPerDay) days := int(duration.Hours() / hoursPerDay)
if days == 1 { if days == 1 {
return "1 day ago" return "1 day ago"
} }
if days < daysPerMonth { if days < daysPerMonth {
return duration.Truncate(hoursPerDay*time.Hour).String() + " ago" return duration.Truncate(hoursPerDay*time.Hour).String() + " ago"
} }
@@ -118,7 +122,7 @@ func initTemplates() {
"appLicense": func() string { return version.License }, "appLicense": func() string { return version.License },
"appRepoURL": func() string { return version.RepoURL }, "appRepoURL": func() string { return version.RepoURL },
"appGitRevision": func() string { return version.GitRevisionShort }, "appGitRevision": func() string { return version.GitRevisionShort },
"appGitCommitURL": func() string { return version.CommitURL() }, "appGitCommitURL": version.CommitURL,
} }
// Parse index template // Parse index template
@@ -134,19 +138,22 @@ func initTemplates() {
} }
// Parse AS detail template // Parse AS detail template
defaultTemplates.ASDetail, err = template.New("asDetail").Funcs(funcs).Parse(asDetailHTML) defaultTemplates.ASDetail, err = template.New("asDetail").
Funcs(funcs).Parse(asDetailHTML)
if err != nil { if err != nil {
panic("failed to parse AS detail template: " + err.Error()) panic("failed to parse AS detail template: " + err.Error())
} }
// Parse prefix detail template // Parse prefix detail template
defaultTemplates.PrefixDetail, err = template.New("prefixDetail").Funcs(funcs).Parse(prefixDetailHTML) defaultTemplates.PrefixDetail, err = template.New("prefixDetail").
Funcs(funcs).Parse(prefixDetailHTML)
if err != nil { if err != nil {
panic("failed to parse prefix detail template: " + err.Error()) panic("failed to parse prefix detail template: " + err.Error())
} }
// Parse prefix length template // Parse prefix length template
defaultTemplates.PrefixLength, err = template.New("prefixLength").Funcs(funcs).Parse(prefixLengthHTML) defaultTemplates.PrefixLength, err = template.New("prefixLength").
Funcs(funcs).Parse(prefixLengthHTML)
if err != nil { if err != nil {
panic("failed to parse prefix length template: " + err.Error()) panic("failed to parse prefix length template: " + err.Error())
} }
internal/version/version.go
+3 -2
View File
@@ -3,11 +3,12 @@ package version
// Build-time variables set via ldflags // Build-time variables set via ldflags
// //
//nolint:gochecknoglobals // These must be variables to allow ldflags injection at build time //nolint:gochecknoglobals // variables so that ldflags can set them at build time
var ( var (
// GitRevision is the git commit hash // GitRevision is the git commit hash
GitRevision = "unknown" GitRevision = "unknown"
// GitRevisionShort is the short git commit hash (7 chars) // GitRevisionShort is the version the page footer shows: the tag or
// short commit hash from `git describe --tags --always`
GitRevisionShort = "unknown" GitRevisionShort = "unknown"
) )
internal/whois/whois.go
+106 -78
View File
@@ -98,7 +98,8 @@ func (c *Client) LookupASN(ctx context.Context, asn int) (*ASNInfo, error) {
if err != nil { if err != nil {
// Return partial data from IANA if RIR query fails // Return partial data from IANA if RIR query fails
info := c.parseResponse(asn, rir, ianaResp) info := c.parseResponse(asn, rir, ianaResp)
info.RawResponse = ianaResp + "\n--- RIR query failed: " + err.Error() + " ---\n" info.RawResponse = fmt.Sprintf("%s\n--- RIR query failed: %v ---\n",
ianaResp, err)
return info, nil return info, nil
} }
@@ -120,28 +121,33 @@ func (c *Client) query(ctx context.Context, server, query string) (string, error
defer func() { _ = conn.Close() }() defer func() { _ = conn.Close() }()
// Set deadlines // Set deadlines
if err := conn.SetWriteDeadline(time.Now().Add(writeTimeout)); err != nil { err = conn.SetWriteDeadline(time.Now().Add(writeTimeout))
if err != nil {
return "", fmt.Errorf("set write deadline: %w", err) return "", fmt.Errorf("set write deadline: %w", err)
} }
// Send query // Send query
if _, err := fmt.Fprintf(conn, "%s\r\n", query); err != nil { _, err = fmt.Fprintf(conn, "%s\r\n", query)
if err != nil {
return "", fmt.Errorf("write query: %w", err) return "", fmt.Errorf("write query: %w", err)
} }
// Read response // Read response
if err := conn.SetReadDeadline(time.Now().Add(readTimeout)); err != nil { err = conn.SetReadDeadline(time.Now().Add(readTimeout))
if err != nil {
return "", fmt.Errorf("set read deadline: %w", err) return "", fmt.Errorf("set read deadline: %w", err)
} }
var sb strings.Builder var sb strings.Builder
scanner := bufio.NewScanner(conn) scanner := bufio.NewScanner(conn)
for scanner.Scan() { for scanner.Scan() {
sb.WriteString(scanner.Text()) sb.WriteString(scanner.Text())
sb.WriteString("\n") sb.WriteString("\n")
} }
if err := scanner.Err(); err != nil { err = scanner.Err()
if err != nil {
return sb.String(), fmt.Errorf("read response: %w", err) return sb.String(), fmt.Errorf("read response: %w", err)
} }
@@ -149,10 +155,10 @@ func (c *Client) query(ctx context.Context, server, query string) (string, error
} }
// parseIANAReferral extracts the RIR and WHOIS server from an IANA response. // parseIANAReferral extracts the RIR and WHOIS server from an IANA response.
func (c *Client) parseIANAReferral(response string) (rir, whoisServer string) { func (c *Client) parseIANAReferral(response string) (string, string) {
lines := strings.Split(response, "\n") var rir string
for _, line := range lines { for line := range strings.SplitSeq(response, "\n") {
line = strings.TrimSpace(line) line = strings.TrimSpace(line)
// Look for "refer:" line // Look for "refer:" line
@@ -183,18 +189,9 @@ func (c *Client) parseIANAReferral(response string) (rir, whoisServer string) {
// Also check organisation line for RIR hints // Also check organisation line for RIR hints
if strings.HasPrefix(strings.ToLower(line), "organisation:") { if strings.HasPrefix(strings.ToLower(line), "organisation:") {
org := strings.ToLower(line) orgRIR := rirInOrganisationLine(strings.ToLower(line))
switch { if orgRIR != "" {
case strings.Contains(org, "arin"): rir = orgRIR
rir = RIRARIN
case strings.Contains(org, "ripe"):
rir = RIRRIPE
case strings.Contains(org, "apnic"):
rir = RIRAPNIC
case strings.Contains(org, "lacnic"):
rir = RIRLACNIC
case strings.Contains(org, "afrinic"):
rir = RIRAFRNIC
} }
} }
} }
@@ -202,6 +199,25 @@ func (c *Client) parseIANAReferral(response string) (rir, whoisServer string) {
return rir, "" return rir, ""
} }
// rirInOrganisationLine returns the RIR a lowercased IANA "organisation:"
// line names, or "" when it names none.
func rirInOrganisationLine(org string) string {
switch {
case strings.Contains(org, "arin"):
return RIRARIN
case strings.Contains(org, "ripe"):
return RIRRIPE
case strings.Contains(org, "apnic"):
return RIRAPNIC
case strings.Contains(org, "lacnic"):
return RIRLACNIC
case strings.Contains(org, "afrinic"):
return RIRAFRNIC
default:
return ""
}
}
// parseResponse extracts ASN information from a WHOIS response. // parseResponse extracts ASN information from a WHOIS response.
func (c *Client) parseResponse(asn int, rir, response string) *ASNInfo { func (c *Client) parseResponse(asn int, rir, response string) *ASNInfo {
info := &ASNInfo{ info := &ASNInfo{
@@ -211,11 +227,13 @@ func (c *Client) parseResponse(asn int, rir, response string) *ASNInfo {
} }
lines := strings.Split(response, "\n") lines := strings.Split(response, "\n")
var addressLines []string var addressLines []string
for _, line := range lines { for _, line := range lines {
line = strings.TrimSpace(line) line = strings.TrimSpace(line)
if line == "" || strings.HasPrefix(line, "%") || strings.HasPrefix(line, "#") { if line == "" || strings.HasPrefix(line, "%") ||
strings.HasPrefix(line, "#") {
continue continue
} }
@@ -232,63 +250,14 @@ func (c *Client) parseResponse(asn int, rir, response string) *ASNInfo {
continue continue
} }
switch key {
// AS Name (varies by RIR)
case "asname", "as-name":
if info.ASName == "" {
info.ASName = value
}
// Organization
case "orgname", "org-name", "owner":
if info.OrgName == "" {
info.OrgName = value
}
case "orgid", "org-id", "org":
if info.OrgID == "" {
info.OrgID = value
}
// Address (collect multiple lines) // Address (collect multiple lines)
case "address": if key == "address" {
addressLines = append(addressLines, value) addressLines = append(addressLines, value)
// Country continue
case "country":
if info.CountryCode == "" && len(value) == 2 {
info.CountryCode = strings.ToUpper(value)
}
// Abuse contact
case "orgabuseemail", "abuse-mailbox":
if info.AbuseEmail == "" {
info.AbuseEmail = value
}
case "orgabusephone":
if info.AbusePhone == "" {
info.AbusePhone = value
}
// Tech contact
case "orgtechemail":
if info.TechEmail == "" {
info.TechEmail = value
}
case "orgtechphone":
if info.TechPhone == "" {
info.TechPhone = value
}
// Registration dates
case "regdate", "created":
if info.RegDate == nil {
info.RegDate = c.parseDate(value)
}
case "updated", "last-modified", "changed":
if info.LastMod == nil {
info.LastMod = c.parseDate(value)
}
} }
c.setResponseField(info, key, value)
} }
// Combine address lines // Combine address lines
@@ -304,6 +273,65 @@ func (c *Client) parseResponse(asn int, rir, response string) *ASNInfo {
return info return info
} }
// setResponseField stores the value of one "key: value" line of a WHOIS
// response in info, keeping the first value seen for each field.
func (c *Client) setResponseField(info *ASNInfo, key, value string) {
switch key {
// Country
case "country":
if info.CountryCode == "" && len(value) == 2 {
info.CountryCode = strings.ToUpper(value)
}
// Registration dates
case "regdate", "created":
if info.RegDate == nil {
info.RegDate = c.parseDate(value)
}
case "updated", "last-modified", "changed":
if info.LastMod == nil {
info.LastMod = c.parseDate(value)
}
default:
field := textFieldForKey(info, key)
if field != nil && *field == "" {
*field = value
}
}
}
// textFieldForKey returns the text field of info that a WHOIS response key
// fills, or nil when the key fills none.
func textFieldForKey(info *ASNInfo, key string) *string {
switch key {
// AS Name (varies by RIR)
case "asname", "as-name":
return &info.ASName
// Organization
case "orgname", "org-name", "owner":
return &info.OrgName
case "orgid", "org-id", "org":
return &info.OrgID
// Abuse contact
case "orgabuseemail", "abuse-mailbox":
return &info.AbuseEmail
case "orgabusephone":
return &info.AbusePhone
// Tech contact
case "orgtechemail":
return &info.TechEmail
case "orgtechphone":
return &info.TechPhone
default:
return nil
}
}
// parseDate attempts to parse various date formats used in WHOIS responses. // parseDate attempts to parse various date formats used in WHOIS responses.
func (c *Client) parseDate(value string) *time.Time { func (c *Client) parseDate(value string) *time.Time {
// Common formats // Common formats
@@ -319,15 +347,15 @@ func (c *Client) parseDate(value string) *time.Time {
value = strings.TrimSpace(value) value = strings.TrimSpace(value)
// Handle "YYYYMMDD" format from LACNIC // Handle "YYYYMMDD" format from LACNIC
if len(value) == lacnicDateFormatLen { if len(value) == lacnicDateFormatLen {
if _, err := time.Parse("20060102", value); err == nil { t, err := time.Parse("20060102", value)
t, _ := time.Parse("20060102", value) if err == nil {
return &t return &t
} }
} }
for _, format := range formats { for _, format := range formats {
if t, err := time.Parse(format, value); err == nil { t, err := time.Parse(format, value)
if err == nil {
return &t return &t
} }
} }
pkg/asinfo/asinfo.go
+4 -2
View File
@@ -1,4 +1,3 @@
// Package asinfo provides information about Autonomous Systems (AS)
package asinfo package asinfo
import ( import (
@@ -54,7 +53,9 @@ func initRegistry() {
} }
var asInfos []Info var asInfos []Info
if err := json.Unmarshal(jsonData, &asInfos); err != nil {
err = json.Unmarshal(jsonData, &asInfos)
if err != nil {
panic("failed to unmarshal embedded AS data: " + err.Error()) panic("failed to unmarshal embedded AS data: " + err.Error())
} }
@@ -137,6 +138,7 @@ func Search(query string) []Info {
defer defaultRegistry.mu.RUnlock() defer defaultRegistry.mu.RUnlock()
var results []Info var results []Info
for _, info := range defaultRegistry.byASN { for _, info := range defaultRegistry.byASN {
if contains(info.Handle, query) || contains(info.Description, query) { if contains(info.Handle, query) || contains(info.Description, query) {
results = append(results, *info) results = append(results, *info)
pkg/asinfo/asinfo_test.go
+52 -16
View File
@@ -1,10 +1,15 @@
package asinfo package asinfo_test
import ( import (
"strings"
"testing" "testing"
"git.eeqj.de/sneak/routewatch/pkg/asinfo"
) )
func TestGet(t *testing.T) { func TestGet(t *testing.T) {
t.Parallel()
tests := []struct { tests := []struct {
name string name string
asn int asn int
@@ -38,20 +43,26 @@ func TestGet(t *testing.T) {
for _, tt := range tests { for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) { t.Run(tt.name, func(t *testing.T) {
info, ok := Get(tt.asn) t.Parallel()
info, ok := asinfo.Get(tt.asn)
if ok != tt.wantOK { if ok != tt.wantOK {
t.Errorf("Get(%d) ok = %v, want %v", tt.asn, ok, tt.wantOK) t.Errorf("Get(%d) ok = %v, want %v", tt.asn, ok, tt.wantOK)
return return
} }
if ok && info.Description != tt.wantDesc { if ok && info.Description != tt.wantDesc {
t.Errorf("Get(%d) description = %q, want %q", tt.asn, info.Description, tt.wantDesc) t.Errorf("Get(%d) description = %q, want %q",
tt.asn, info.Description, tt.wantDesc)
} }
}) })
} }
} }
func TestGetDescription(t *testing.T) { func TestGetDescription(t *testing.T) {
t.Parallel()
tests := []struct { tests := []struct {
asn int asn int
want string want string
@@ -62,7 +73,10 @@ func TestGetDescription(t *testing.T) {
for _, tt := range tests { for _, tt := range tests {
t.Run("", func(t *testing.T) { t.Run("", func(t *testing.T) {
if got := GetDescription(tt.asn); got != tt.want { t.Parallel()
got := asinfo.GetDescription(tt.asn)
if got != tt.want {
t.Errorf("GetDescription(%d) = %q, want %q", tt.asn, got, tt.want) t.Errorf("GetDescription(%d) = %q, want %q", tt.asn, got, tt.want)
} }
}) })
@@ -70,6 +84,8 @@ func TestGetDescription(t *testing.T) {
} }
func TestGetHandle(t *testing.T) { func TestGetHandle(t *testing.T) {
t.Parallel()
tests := []struct { tests := []struct {
asn int asn int
want string want string
@@ -81,7 +97,10 @@ func TestGetHandle(t *testing.T) {
for _, tt := range tests { for _, tt := range tests {
t.Run("", func(t *testing.T) { t.Run("", func(t *testing.T) {
if got := GetHandle(tt.asn); got != tt.want { t.Parallel()
got := asinfo.GetHandle(tt.asn)
if got != tt.want {
t.Errorf("GetHandle(%d) = %q, want %q", tt.asn, got, tt.want) t.Errorf("GetHandle(%d) = %q, want %q", tt.asn, got, tt.want)
} }
}) })
@@ -89,18 +108,23 @@ func TestGetHandle(t *testing.T) {
} }
func TestTotal(t *testing.T) { func TestTotal(t *testing.T) {
total := Total() t.Parallel()
total := asinfo.Total()
if total < 100000 { if total < 100000 {
t.Errorf("Total() = %d, expected > 100000", total) t.Errorf("Total() = %d, expected > 100000", total)
} }
// Verify it's consistent // Verify it's consistent
if total2 := Total(); total2 != total { total2 := asinfo.Total()
if total2 != total {
t.Errorf("Total() returned different values: %d vs %d", total, total2) t.Errorf("Total() returned different values: %d vs %d", total, total2)
} }
} }
func TestSearch(t *testing.T) { func TestSearch(t *testing.T) {
t.Parallel()
tests := []struct { tests := []struct {
name string name string
query string query string
@@ -130,16 +154,21 @@ func TestSearch(t *testing.T) {
for _, tt := range tests { for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) { t.Run(tt.name, func(t *testing.T) {
results := Search(tt.query) t.Parallel()
results := asinfo.Search(tt.query)
if len(results) < tt.wantMin { if len(results) < tt.wantMin {
t.Errorf("Search(%q) returned %d results, want at least %d", tt.query, len(results), tt.wantMin) t.Errorf("Search(%q) returned %d results, want at least %d",
tt.query, len(results), tt.wantMin)
} }
// Verify all results contain the query // Verify all results contain the query
if tt.query != "" { if tt.query != "" {
for _, r := range results { for _, r := range results {
if !contains(r.Handle, tt.query) && !contains(r.Description, tt.query) { if !strings.Contains(r.Handle, tt.query) &&
t.Errorf("Search result ASN %d doesn't contain query %q", r.ASN, tt.query) !strings.Contains(r.Description, tt.query) {
t.Errorf("Search result ASN %d doesn't contain query %q",
r.ASN, tt.query)
} }
} }
} }
@@ -148,13 +177,17 @@ func TestSearch(t *testing.T) {
} }
func TestDataIntegrity(t *testing.T) { func TestDataIntegrity(t *testing.T) {
t.Parallel()
// Verify no duplicate ASNs // Verify no duplicate ASNs
all := All() all := asinfo.All()
seen := make(map[int]bool) seen := make(map[int]bool)
for _, info := range all { for _, info := range all {
if seen[info.ASN] { if seen[info.ASN] {
t.Errorf("Duplicate ASN found: %d", info.ASN) t.Errorf("Duplicate ASN found: %d", info.ASN)
} }
seen[info.ASN] = true seen[info.ASN] = true
} }
@@ -163,6 +196,7 @@ func TestDataIntegrity(t *testing.T) {
if info.Handle == "" && info.ASN != 0 { if info.Handle == "" && info.ASN != 0 {
t.Errorf("ASN %d has empty handle", info.ASN) t.Errorf("ASN %d has empty handle", info.ASN)
} }
if info.Description == "" { if info.Description == "" {
t.Errorf("ASN %d has empty description", info.ASN) t.Errorf("ASN %d has empty description", info.ASN)
} }
@@ -174,8 +208,9 @@ func BenchmarkGet(b *testing.B) {
asns := []int{1, 15169, 13335, 32934, 8075, 16509} asns := []int{1, 15169, 13335, 32934, 8075, 16509}
b.ResetTimer() b.ResetTimer()
for i := 0; i < b.N; i++ {
Get(asns[i%len(asns)]) for i := range b.N {
asinfo.Get(asns[i%len(asns)])
} }
} }
@@ -183,7 +218,8 @@ func BenchmarkSearch(b *testing.B) {
queries := []string{"Google", "Amazon", "Microsoft", "University"} queries := []string{"Google", "Amazon", "Microsoft", "University"}
b.ResetTimer() b.ResetTimer()
for i := 0; i < b.N; i++ {
Search(queries[i%len(queries)]) for i := range b.N {
asinfo.Search(queries[i%len(queries)])
} }
} }
script/bootstrap
+6 -6
View File
@@ -3,7 +3,8 @@
# this repo. Idempotent: every install is guarded by a check so already # this repo. Idempotent: every install is guarded by a check so already
# installed tools are skipped. Base tooling comes from nix, apt, brew, # installed tools are skipped. Base tooling comes from nix, apt, brew,
# or apk (detected in that order); assumes NOTHING is present (not git, # or apk (detected in that order); assumes NOTHING is present (not git,
# make, or go). # make, or go). The linter and the tests run only in Docker, as phases of
# the Dockerfile, so nothing installs golangci-lint here.
set -eu set -eu
ROOT="$(cd "$(dirname "$0")/.." && pwd -P)" ROOT="$(cd "$(dirname "$0")/.." && pwd -P)"
@@ -58,11 +59,10 @@ main() {
# Go toolchain # Go toolchain
if missing go; then pkg_install go golang go go; fi if missing go; then pkg_install go golang go go; fi
# golangci-lint: packaged in nix, brew, and apk. There is no apt # Warn, don't fail: make build works without it.
# package; on apt systems install it manually from a hash-verified if missing docker; then
# GitHub release archive (never curl | sh). echo "bootstrap: WARNING: docker not found; install it to run" \
if missing golangci-lint; then "make test, make lint, make check and make docker." >&2
pkg_install golangci-lint golangci-lint golangci-lint golangci-lint
fi fi
go mod download go mod download
script/check
+3 -2
View File
@@ -1,7 +1,8 @@
#!/bin/sh #!/bin/sh
# script/check: run all checks (test, lint, fmt-check). Our own # script/check: run all checks (test, lint, fmt-check). Our own
# extension to scripts-to-rule-them-all. Must not modify any files. # extension to scripts-to-rule-them-all. test and lint are Docker
# Generic: usually needs no adaptation. # phases; fmt-check is native, because a formatter writes the working
# tree. Must not modify any files.
set -eu set -eu
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)" SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
script/cibuild
+19 -5
View File
@@ -1,14 +1,28 @@
#!/bin/sh #!/bin/sh
# script/cibuild: run the CI build. The Dockerfile runs script/check # script/cibuild: run the CI build. It bootstraps first: a CI runner
# (via make check), so a successful build implies all checks pass. # checks out and runs this and nothing else, and script/fmt-check runs
# Generic: needs no adaptation. The Gitea workflow runs this on push. # the formatter on the host, which a pristine checkout cannot do.
# --no-cache for the same reason as script/docker: the gate phases the
# final stage depends on are RUN steps, and a cached one is a check that
# did not run.
set -eu set -eu
ROOT="$(cd "$(dirname "$0")/.." && pwd -P)" SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
ROOT="$(cd "$SCRIPT_DIR/.." && pwd -P)"
main() { main() {
cd "$ROOT" cd "$ROOT"
docker build . "$SCRIPT_DIR/bootstrap"
"$SCRIPT_DIR/check"
# Own line: a failing command substitution inside an argument does
# not trip `set -e`, so the inline form degrades silently to an
# empty constant. The VERSION build argument takes precedence over
# the version a build stage derives from the .git in the context.
version="$(git describe --tags --always --dirty 2>/dev/null || true)"
[ -n "$version" ] || version="unknown"
docker build --no-cache \
--build-arg VERSION="$version" \
-t "$("$SCRIPT_DIR/projectname")" .
} }
main "$@" main "$@"
script/docker
+11 -2
View File
@@ -1,7 +1,8 @@
#!/bin/sh #!/bin/sh
# script/docker: build the Docker image tagged with the project name. # script/docker: build the Docker image tagged with the project name.
# Identical in all repos; the tag comes from script/projectname. # Identical in all repos; the tag comes from script/projectname.
# Generic: needs no adaptation. # --no-cache because the gate phases the final stage depends on are RUN
# steps, and a cached one is a check that did not run.
set -eu set -eu
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)" SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
@@ -9,7 +10,15 @@ ROOT="$(cd "$SCRIPT_DIR/.." && pwd -P)"
main() { main() {
cd "$ROOT" cd "$ROOT"
docker build -t "$("$SCRIPT_DIR/projectname")" . # Own line: a failing command substitution inside an argument does
# not trip `set -e`, so the inline form degrades silently to an
# empty constant. The VERSION build argument takes precedence over
# the version a build stage derives from the .git in the context.
version="$(git describe --tags --always --dirty 2>/dev/null || true)"
[ -n "$version" ] || version="unknown"
docker build --no-cache \
--build-arg VERSION="$version" \
-t "$("$SCRIPT_DIR/projectname")" .
} }
main "$@" main "$@"
script/lint
+14 -4
View File
@@ -1,13 +1,23 @@
#!/bin/sh #!/bin/sh
# script/lint: run the linters. # script/lint: run the linter. Linting is a phase of the Dockerfile and
# this builds that phase alone; the linter is never installed or run on
# a developer host, where a shared result cache and a host-global lock
# make its answer untrustworthy.
#
# The phase is not the last stage in the file, so it is built only when
# --target names it. --no-cache because a cached lint layer is a lint
# that did not run. The tag makes each build replace the previous image
# instead of leaving a dangling one behind.
set -eu set -eu
ROOT="$(cd "$(dirname "$0")/.." && pwd -P)" SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
ROOT="$(cd "$SCRIPT_DIR/.." && pwd -P)"
main() { main() {
cd "$ROOT" cd "$ROOT"
go vet ./... docker build --no-cache \
golangci-lint run --target lint \
-t "$("$SCRIPT_DIR/projectname")-lint" .
} }
main "$@" main "$@"
script/test
+10 -8
View File
@@ -1,17 +1,19 @@
#!/bin/sh #!/bin/sh
# script/test: run the test suite. On failure, rerun verbosely so the # script/test: run the test suite. Testing is a phase of the Dockerfile
# failing tests are visible in the output. # and this builds that phase alone, on the same terms as script/lint:
# --target because a phase that is not the last stage is built only when
# named, --no-cache because a cached test layer is a test that did not
# run, and a tag so each build replaces the previous image.
set -eu set -eu
ROOT="$(cd "$(dirname "$0")/.." && pwd -P)" SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
ROOT="$(cd "$SCRIPT_DIR/.." && pwd -P)"
main() { main() {
cd "$ROOT" cd "$ROOT"
go test -short -timeout 30s -race -cover ./... || { docker build --no-cache \
echo "--- Rerunning with -v for details ---" --target test \
go test -short -timeout 30s -race -v ./... -t "$("$SCRIPT_DIR/projectname")-test" .
exit 1
}
} }
main "$@" main "$@"