The repository had no licence file at all, which makes publicly readable
code all-rights-reserved by default: nobody may legally use it. That is a
1.0 blocker rather than a nicety, and `LICENSE` was the only file from
`REPO_POLICIES.md`'s required minimum still missing here.
The choice is standing org policy rather than a per-repo call: any public
repo lacking a licence gets MIT, while a private repo with no licence is
already all-rights-reserved and needs nothing. `sneak/dnswatcher` is
public, so MIT.
`LICENSE` carries the canonical MIT text byte-for-byte with only the
copyright line filled in; no clauses added, removed, reworded, or
reflowed. `README.md`'s first line now names the licence, which the
Description requirement in `REPO_POLICIES.md` calls for, and the License
section states MIT and points at the file instead of recording the
decision as pending.
`script/test` did not pass `-count=1`, so on an unchanged tree Go
served the whole suite from its test cache: exit 0 in ~0.2s with every
package marked `(cached)` and not one DNS query made. This repo's suite
exists to exercise live resolution on every run (`TESTING.md`), so that
green asserted nothing — and it is exactly the green used as evidence
that a flakiness fix works, since "run it a few times" stops being
runs after the first.
`-count=1` now disables caching on every invocation.
The conditional verbose rerun that `REPO_POLICIES.md` mandates was
missing at the same spot and is added here rather than left broken: the
primary run had been unconditionally `-v`, which is the failure mode
the policy exists to prevent (unreadable CI and `docker build` logs on
success). Tests now run quiet, and only a failure triggers the `-v`
rerun. The rerun carries `-count=1` too, so it cannot replay a cached
copy of the failure it is meant to diagnose, and its exit status is
discarded in favour of a forced 1: the first failure already proved the
suite broken, so a flake that passes the second time must not turn the
build green.
`-timeout 90s` is untouched. It is a deliberate backstop that must
strictly exceed the 60s hard cap on suite duration.
No special-casing for the Docker build, which also reaches this script
via `RUN make test`: a fresh container's test cache is empty, so
`-count=1` changes nothing there and carving out an exception would
only create a second code path that could drift.
Verified: three back-to-back `make test` runs on an unchanged tree,
zero `(cached)` markers, ~4.0-4.5s wall each (was ~0.2s cached),
comfortably inside the 20s target with `-race` and `-cover` both still
working and coverage percentages unchanged. The rerun-and-still-fail
path was exercised against a purpose-built flaky test that fails once
then passes: quiet failure, verbose rerun that genuinely re-executed,
exit 1 regardless. `make check` green.
Rework of the unit at #93
(commit 9cb2c2b), against the review at
#136 (comment).
Review of 9cb2c2b found the quorum assertions could not fail on a
class of wrong answer. Each test banned exactly one bad status —
_AllReturnOK banned only nxdomain, _NXDomainFromAllNS banned only ok
— so resolver.StatusNoData passed both. nodata is a wrong answer, not
silence, and answeredCount counted it as answered, so it did not even
trigger a retry; with a quorum of 3 of 4 a single wrong nameserver
slid through undetected. That is assertion-loosening beyond what the
quorum change requires.
Tolerance is now a closed allowlist rather than a blocklist of one
status. unsanctionedStatuses() reports every per-nameserver result
whose status the caller did not explicitly sanction: ok/timeout/error
for the all-OK test, nxdomain/timeout/error for the NXDOMAIN test.
Silence (timeout, error) is the only thing quorum exists to tolerate;
any other status, including one added to the resolver later, fails by
name. answeredCount is likewise an allowlist of ok/nxdomain/nodata, so
an unknown status counts as silence and can only cause a retry and
then a loud failure, never a quiet pass.
Two harness tests cover the regression directly: three OK plus one
nodata (quorum satisfied, no nxdomain present — the input that used
to pass) is now reported as unsanctioned, and an unknown status is
neither counted as answered nor tolerated.
Verified by re-running the reviewer's probe: queryEachNS patched to
force one of google.com's four nameservers to return StatusNoData
turns both tests red, naming the offending nameserver and status —
--- FAIL: TestQueryAllNameservers_AllReturnOK (1.12s)
Should be empty, but was [ns1.google.com.=nodata]
every nameserver must answer OK or not answer at all:
ns1.google.com.=nodata ns2.google.com.=ok ns3.google.com.=ok
ns4.google.com.=ok
--- FAIL: TestQueryAllNameservers_NXDomainFromAllNS (1.34s)
Should be empty, but was [ns1.google.com.=nodata]
every nameserver must report NXDOMAIN or not answer at all:
ns1.google.com.=nodata ns2.google.com.=nxdomain
ns3.google.com.=nxdomain ns4.google.com.=nxdomain
— and green with the probe reverted. Also fixes the review's nit: the
per-attempt deadline assertion had no lower bound, so it passed for a
deadline far shorter than intended.
No production code changed; DNS is still never mocked.
The resolver's live-DNS tests failed nondeterministically, a different
subset each run. Three structural causes, all test-side:
- Burst fan-out. Every test in the package is parallel and the build
hosts have many cores, so all ~35 iterative resolutions started at
the same instant and, because queryServers walks rootServerList() in
fixed order, hit the same root server within milliseconds. Root
servers rate-limit that.
- No retry anywhere. One dropped UDP packet in a delegation chain
failed a test outright.
- Unanimity assertions. TestQueryAllNameservers_AllReturnOK and
_NXDomainFromAllNS required every one of a domain's nameservers to
answer, with no tolerance for one being slow.
New internal/resolver/livedns_test.go addresses each: a package-wide
gate bounds how many live resolutions are in flight at once, every
live operation gets three attempts with exponential backoff and its
own deadline, and multi-nameserver assertions now need a strict
majority rather than unanimity. The retry predicate is deliberately
transport-level -- "did a nameserver answer at all" -- never the
assertion under test, so a resolver that answers incorrectly still
fails on the first attempt. A nameserver that stays silent is
tolerated; one that answers wrongly is not.
livedns_harness_test.go tests that machinery directly: quorum
arithmetic, status counting, the gate's concurrency bound, per-attempt
deadlines, and recovery from a transient failure. It touches no DNS.
Nothing is mocked, faked, stubbed, recorded, skipped or build-tagged,
and production resolver behaviour is unchanged.
Test caps move to the new org-wide values ruled at prompts issue 41:
60s hard cap, 20s target, 90s -timeout backstop. REPO_POLICIES.md is
re-vendored byte-identical from sneak/prompts rather than hand-edited,
which also picks up the golangci-lint paragraph this copy had drifted
behind on. TESTING.md's stale 30-second target follows to 60.
#93
golangci-lint is no longer installed or run on the host. script/lint is
now a thin wrapper that builds the new root Dockerfile.lint, which COPYs
the repo into the digest-pinned golangci/golangci-lint:v2.12.2 image and
lints as a build step, so a successful build is a clean lint. This works
even where the docker daemon is remote and bind mounts are impossible.
Dockerfile.lint is split into a deps stage (base image, go mod download)
and a lint stage (source copy, linter run). script/lint passes
--no-cache-filter=lint so the lint stage executes on every invocation:
caching is explicitly waived for linting, and a cached build lints
nothing. The deps stage stays cached and no global cache invalidation is
performed. --progress=plain keeps the linter's own output visible.
golangci-lint config verify is deliberately omitted: it fetches its JSON
schema over a live, unpinned HTTPS call, which would make linting
network-dependent and defeat hash-pinning.
script/bootstrap no longer installs golangci-lint and warns instead when
docker is absent. The goimports install stays, since script/fmt and
script/fmt-check still run it on the host.
The root Dockerfile ran make check in its builder stage, which would now
recurse into script/lint and shell out to docker build with no daemon
available. It gains its own lint stage on the same pinned image, invoked
directly, with the builder depending on it via COPY --from=lint and
running make fmt-check, make test and make build.
Updates golangci-lint to v2.12.2 and sets `.golangci.yml` to the org-standard v2-schema config already deployed across the org's repos. The config change is owner-authorized (see #96 (comment) and #96 (comment)); the same file is being landed as canonical via prompts PR #24 (sneak/prompts#24).
## Changes
- **Commit-pinned installs**: golangci-lint pinned to commit `c0d3ddc9cf3faa61a4e378e879ece580256d76e5` (v2.12.2, released 2026-05-06) in `Dockerfile` and `script/bootstrap`.
- **`.golangci.yml` set to the org-standard v2 config** (sha256 `021cc83f4e6fc7c31b95b34b846723dfcf20b66b7baeea1dc40406e643346bcb`), byte-identical to the file used across the org's other repos. Settings live under `linters.settings`, so the `lll`/`funlen`/`cyclop`/`dupl` thresholds are actually applied (under the old hybrid file, v2 silently ignored the top-level `linters-settings` block).
- **Lint fixes** required by the now-active thresholds:
- `goconst`: shared constants for repeated status/priority/DNS-fixture strings in `internal/watcher/watcher.go` and the notify, state, and watcher tests
- `dupl`: consolidated duplicated ntfy/slack HTTP-error tests and SendNotification endpoint-error tests behind shared helpers in `internal/notify/delivery_test.go`
- `lll`: wrapped long test table entries and comments in `internal/config/classify_test.go`, `internal/notify/history_test.go`, `internal/state/state_test.go`, `internal/watcher/watcher_test.go`; shortened one inline nolint justification in `internal/notify/retry.go`
- **`TODO.md`**: Completed Steps entry updated in the same commit.
- Rebased onto current `main` (`f79cd98`); the branch is one clean commit.
## Notes
- v2.12 deprecates the `gomodguard` linter in favor of `gomodguard_v2`. The org-standard config does not disable the deprecated linter, so golangci-lint may emit an informational deprecation warning; this is accepted by the owner and does not affect the exit status (this exact config+code combination was CI-green at `dea7e44`).
## Verification
- `make check` exits 0 (fmt-check, tests, lint)
- `make lint`: 0 issues; no deprecation warning surfaced in the runs performed
- sha256 of `.golangci.yml` at HEAD verified equal to `021cc83f4e6fc7c31b95b34b846723dfcf20b66b7baeea1dc40406e643346bcb`
Co-authored-by: sneak <sneak@sneak.berlin>
Reviewed-on: #96
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
Adds a prominent "No DNS mocking. Ever." section near the top of `README.md`, per owner policy (sneak, 2026-08-07):
- DNS is never mocked in this project — no mock resolvers, fake DNS servers, or stubbed lookups, in tests or anywhere else.
- Tests exercise real iterative resolution against live nameservers by design.
- Flaky live tests are fixed with robustness (retries, multiple nameservers, timeouts) or explicit opt-in gating decided by the owner — never with mocks.
- Contributions introducing DNS mocks will be rejected.
Markdown-only change; matches the README's existing tone and hard-wrap style. `script/fmt` covers Go only, so no formatter output applies to this file. Verified via `script/cibuild` (docker build runs `make check` with the pinned toolchain) — green. A direct local `make check` shows 21 pre-existing `goconst` lint findings that come from a newer local `golangci-lint` (v2.12.2 vs the pinned v2.10.1) and are unrelated to this change.
Related: #93 is being reframed under this policy.
Co-authored-by: sneak <sneak@sneak.berlin>
Reviewed-on: #95
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Notifications were fire-and-forget: if Slack, Mattermost, or ntfy was temporarily down, changes were silently lost. This adds automatic retry with exponential backoff and jitter to all notification endpoints.
## Changes
### New file: `internal/notify/retry.go`
- `RetryConfig` struct with configurable max retries, base delay, max delay
- `backoff()` computes delay as `BaseDelay * 2^attempt`, capped at `MaxDelay`, with ±25% jitter
- `deliverWithRetry()` wraps any send function with the retry loop
- Defaults: 3 retries (4 total attempts), 1s base delay, 10s max delay
- Context-aware: respects cancellation during retry sleep
- Injectable `sleepFn` for test determinism
### Modified: `internal/notify/notify.go`
- Added `retryConfig` and `sleepFn` fields to `Service`
- Updated `dispatchNtfy`, `dispatchSlack`, `dispatchMattermost` to wrap sends in `deliverWithRetry`
- Structured logging: warns on each retry, logs error only after all retries exhausted, logs info on success after retry
### Modified: `internal/notify/export_test.go`
- Added test helpers: `SetRetryConfig`, `SetSleepFunc`, `DeliverWithRetry`, `BackoffDuration`
### New file: `internal/notify/retry_test.go`
- Backoff calculation tests (exponential increase, max cap with jitter)
- `deliverWithRetry` unit tests: first-attempt success, transient failure recovery, exhausted retries, context cancellation
- Integration tests via `SendNotification`: transient failure retries, all-endpoints retry independently, permanent failure exhausts retries
## Verification
- `make fmt` ✅
- `make check` (format + lint + tests + build) ✅
- `docker build .` ✅
- All existing tests continue to pass unchanged
- No DNS client mocking — notification tests use `httptest` servers
closes #62
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Reviewed-on: #87
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
Closes [issue #88](#88).
Changes the default `DNSWATCHER_DATA_DIR` from the relative path `./data` to the absolute path `/var/lib/dnswatcher`, following the [Filesystem Hierarchy Standard](https://refspecs.linuxfoundation.org/FHS_3.0/fhs/ch05s08.html) convention for variable application state data.
## Changes
- **`internal/config/config.go`**: Changed the Viper default for `DATA_DIR` from `"./data"` to `"/var/lib/"+name`, where `name` is the application name ("dnswatcher"). This makes the default derived from the app name rather than hardcoded.
- **`internal/config/config_test.go`**: Updated `TestNew_DefaultValues` and `TestStatePath` to expect the new absolute default.
- **`README.md`**: Updated the environment variable table and `.env` example to show `/var/lib/dnswatcher` as the default.
The Dockerfile already set `ENV DNSWATCHER_DATA_DIR=/var/lib/dnswatcher` explicitly, so Docker deployments are unaffected. This change makes the code default consistent with the Docker configuration.
`docker build .` passes all checks (fmt, lint, tests, build).
Co-authored-by: user <user@Mac.lan guest wan>
Reviewed-on: #89
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Enhances the `/api/v1/status` endpoint to return comprehensive monitoring state instead of just `{"status": "ok"}`.
## Changes
The endpoint now returns:
- **Summary counts**: domains, hostnames, ports (total + open), certificates (total + ok + error)
- **Domains**: each monitored domain with its discovered nameservers and last check timestamp
- **Hostnames**: each monitored hostname with per-nameserver DNS records, status, and last check timestamps
- **Ports**: each monitored IP:port with open/closed state, associated hostnames, and last check timestamp
- **Certificates**: each TLS certificate with CN, issuer, expiry, SANs, status, and last check timestamp
- **Last updated**: timestamp of the overall monitoring state
All data is derived from the existing `state.GetSnapshot()`, consistent with how the dashboard works. No configuration details (webhook URLs, API tokens) are exposed.
## Example response structure
```json
{
"status": "ok",
"lastUpdated": "2026-03-10T12:00:00Z",
"counts": {
"domains": 2,
"hostnames": 3,
"ports": 10,
"portsOpen": 8,
"certificates": 4,
"certificatesOk": 3,
"certificatesError": 1
},
"domains": { ... },
"hostnames": { ... },
"ports": { ... },
"certificates": { ... }
}
```
closes #73
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Reviewed-on: #86
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
When set to a truthy value, sends a startup status notification to all configured notification channels after the first full scan completes on application startup. The notification is clearly an all-ok/success message showing the number of monitored domains, hostnames, ports, and certificates.
Changes:
- Added `SendTestNotification` config field reading `DNSWATCHER_SEND_TEST_NOTIFICATION`
- Added `maybeSendTestNotification()` in watcher, called after initial `RunOnce` in `Run`
- Added 3 watcher tests (enabled via Run, enabled via RunOnce alone, disabled)
- Added config tests for the new field
- Updated README: env var table, example .env, Docker example
Closes#84
Co-authored-by: user <user@Mac.lan guest wan>
Reviewed-on: #85
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Adds a read-only web dashboard at `GET /` that shows the current monitoring state and recent alerts. Unauthenticated, single-page, no navigation.
## What it shows
- **Summary bar**: counts of monitored domains, hostnames, ports, certificates
- **Domains**: nameservers with last-checked age
- **Hostnames**: per-nameserver DNS records, status badges, relative age
- **Ports**: open/closed state with associated hostnames and age
- **TLS Certificates**: CN, issuer, expiry (color-coded by urgency), status, age
- **Recent Alerts**: last 100 notifications in reverse chronological order with priority badges
Every data point displays its age (e.g. "5m ago") so freshness is visible at a glance. Auto-refreshes every 30 seconds.
## What it does NOT show
No secrets: webhook URLs, ntfy topics, Slack/Mattermost endpoints, API tokens, and configuration details are never exposed.
## Design
All assets (CSS) are embedded in the binary and served from `/s/`. Zero external HTTP requests at runtime — no CDN dependencies or third-party resources. Dark, technical aesthetic with saturated teals and blues on dark slate. Single page — everything on one screen.
## Implementation
- `internal/notify/history.go` — thread-safe ring buffer (`AlertHistory`) storing last 100 alerts
- `internal/notify/notify.go` — records each alert in history before dispatch; refactored `SendNotification` into smaller `dispatch*` helpers to satisfy funlen
- `internal/handlers/dashboard.go` — `HandleDashboard()` handler with embedded HTML template, helper functions (`relTime`, `formatRecords`, `expiryDays`, `joinStrings`)
- `internal/handlers/templates/dashboard.html` — Tailwind-styled single-page dashboard
- `internal/handlers/handlers.go` — added `State` and `Notify` dependencies via fx
- `internal/server/routes.go` — registered `GET /` route
- `static/` — embedded CSS assets served via `/s/` prefix
- `README.md` — documented the dashboard and new endpoint
## Tests
- `internal/notify/history_test.go` — empty, add+recent ordering, overflow beyond capacity
- `internal/handlers/dashboard_test.go` — `relTime`, `expiryDays`, `formatRecords`
- All existing tests pass unchanged
- `docker build .` passes
closes [#82](#82)
<!-- session: rework-pr-83 -->
Co-authored-by: user <user@Mac.lan guest wan>
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Reviewed-on: #83
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Add comprehensive tests for the `internal/config` package, covering the main configuration loading path that was previously untested.
Closes [issue #72](#72)
## What Changed
Added three new test files:
- **`config_test.go`** — 16 tests covering `New()`, `StatePath()`, and the full config loading pipeline
- **`parsecsv_test.go`** — 10 test cases for `parseCSV()` edge cases
- **`export_test.go`** — standard Go export bridge for testing unexported `parseCSV`
## Test Coverage
| Area | Tests |
|------|-------|
| Default values | All 14 config fields verified against documented defaults |
| Environment overrides | All env vars tested including `PORT` (unprefixed) |
| Invalid duration fallback | `DNSWATCHER_DNS_INTERVAL=banana` falls back to 1h |
| Invalid TLS interval | `DNSWATCHER_TLS_INTERVAL=notaduration` falls back to 12h |
| No targets error | Empty/missing `DNSWATCHER_TARGETS` returns `ErrNoTargets` |
| Invalid targets | Public suffix (`co.uk`) rejected with error |
| CSV parsing | Trailing commas, leading commas, consecutive commas, whitespace, tabs |
| Debug mode | `DNSWATCHER_DEBUG=true` enables debug logging |
| Target classification | Domains vs hostnames correctly separated via PSL |
| StatePath | Path construction with various `DataDir` values |
| Empty appname | Falls back to "dnswatcher" config file name |
**Coverage: 23% → 92.5%**
## Notes
- Tests use `viper.Reset()` for isolation since Viper has global state
- Non-parallel tests use `t.Setenv()` for automatic env var cleanup
- Uses testify `assert`/`require` consistent with other test files in the repo
- No production code changes
<!-- session: agent:sdlc-manager:subagent:d7fe6cf2-4746-4793-a738-9df8f5f5f0c6 -->
Co-authored-by: user <user@Mac.lan guest wan>
Reviewed-on: #81
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Fixes documentation inaccuracies in README.md identified during QA audit.
### Changes
**API table (closes #67):**
- Removed `GET /api/v1/domains` and `GET /api/v1/hostnames` from the HTTP API table. These endpoints are not implemented — the only routes in `internal/server/routes.go` are `/health`, `/api/v1/status`, and `/metrics` (conditional).
**Feature claims (closes #68):**
- Removed "Inconsistency resolved" from hostname monitoring features. `detectInconsistencies()` detects current inconsistencies but has no state tracking to detect when they resolve.
- Removed `nxdomain` and `nodata` from the state status values table. While the resolver defines these constants, `buildHostnameState()` in the watcher only ever sets status to `"ok"`. Failed queries set `"error"` via the NS disappearance path. These values are never written to state.
- Removed "Empty response" (NODATA/NXDOMAIN) detection claim. Changes are caught generically by `detectRecordChanges()`, not with specific NODATA/NXDOMAIN labeling.
### What was NOT changed
- "Inconsistency detected" remains — this IS implemented in `detectInconsistencies()`.
- All other feature claims were verified against the code and are accurate.
- No Go source code was modified.
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Co-authored-by: Jeffrey Paul <sneak@noreply.example.org>
Reviewed-on: #74
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
When `DNSWATCHER_TARGETS` is empty (the default), dnswatcher previously started successfully and ran indefinitely monitoring nothing. This is a common misconfiguration — forgetting to set the variable or making a typo in its name — and gave no indication anything was wrong.
## Changes
- Added `ErrNoTargets` sentinel error in `internal/config/config.go`
- Extracted `parseAndValidateTargets()` helper to validate that at least one domain or hostname is configured after target classification
- If no targets are configured, dnswatcher now exits with a clear error: `"no monitoring targets configured: set DNSWATCHER_TARGETS environment variable"`
- Updated README.md to document that `DNSWATCHER_TARGETS` is required and dnswatcher will refuse to start without it
## How it works
The validation runs during config construction (via uber/fx), before the watcher or any other component starts. If `DNSWATCHER_TARGETS` is empty or contains only whitespace/empty entries, `buildConfig()` returns `ErrNoTargets`, which causes fx to fail startup with a clear error message.
This is fail-fast behavior: a monitoring daemon with nothing to monitor is a misconfiguration and should not silently run.
Closes #69
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Reviewed-on: #75
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
The `OnStart` hook previously derived the watcher's context from the fx startup context (`startCtx`) via `context.WithoutCancel()`. While `WithoutCancel` strips cancellation and deadline, using `context.Background()` makes the intent explicit: the watcher's monitoring loop must outlive the fx startup phase and is controlled solely by the `cancel` func called in `OnStop`.
## Changes
- Replace `context.WithCancel(context.WithoutCancel(startCtx))` with `context.WithCancel(context.Background())`
- Add explanatory comment documenting why the watcher context is not derived from the startup context
- Unused `startCtx` parameter changed to `_`
Closes#53
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Co-authored-by: Jeffrey Paul <sneak@noreply.example.org>
Reviewed-on: #63
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Port state keys are `ip:port` with a single `hostname` field. When multiple hostnames resolve to the same IP (shared hosting, CDN), only one hostname was associated. This caused orphaned port state when that hostname removed the IP from DNS while the IP remained valid for other hostnames.
## Changes
### State (`internal/state/state.go`)
- `PortState.Hostname` (string) → `PortState.Hostnames` ([]string)
- Custom `UnmarshalJSON` for backward compatibility: reads old single `hostname` field and migrates to a single-element `hostnames` slice
- Added `DeletePortState` and `GetAllPortKeys` methods for cleanup
### Watcher (`internal/watcher/watcher.go`)
- Refactored `checkAllPorts` into three phases:
1. Build IP:port → hostname associations from current DNS data
2. Check each unique IP:port once with all associated hostnames
3. Clean up stale port state entries with no hostname references
- Port change notifications now list all associated hostnames (`Hosts:` instead of `Host:`)
- Added `buildPortAssociations`, `parsePortKey`, and `cleanupStalePorts` helper functions
### README
- Updated state file format example: `hostname` → `hostnames` (array)
- Updated notification description to reflect multiple hostnames
## Backward Compatibility
Existing state files with the old single `hostname` string are handled gracefully via custom JSON unmarshaling — they are read as single-element `hostnames` slices.
Closes #55
Co-authored-by: clawbot <clawbot@noreply.eeqj.de>
Reviewed-on: #65
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
DNS checks now always complete before port or TLS checks begin, ensuring those checks use freshly resolved IP addresses instead of potentially stale ones from a previous cycle.
## Problem
Port and TLS checks read IP addresses from state that was populated during the most recent DNS check. If DNS changes between cycles, port/TLS checks may target stale IPs. In particular, when the TLS ticker fired (every 12h), it ran `runTLSChecks` without refreshing DNS first — meaning TLS checks could use IPs that were up to 12 hours old.
## Changes
- **Extract `runDNSChecks()`** from the former `runDNSAndPortChecks()` so DNS resolution can be invoked independently as a prerequisite for any check type.
- **TLS ticker now runs DNS first**: When the TLS ticker fires, DNS checks run before TLS checks, ensuring fresh IPs.
- **`RunOnce` uses explicit 3-phase ordering**: DNS → ports → TLS. Port checks must complete before TLS because TLS checks only target IPs where port 443 is open.
- **New test `TestDNSRunsBeforePortAndTLSChecks`**: Verifies that when DNS IPs change between cycles, port and TLS checks pick up the new IPs.
- **README updated**: Monitoring lifecycle section now documents the DNS-first ordering guarantee.
## Check ordering
| Trigger | Phase 1 | Phase 2 | Phase 3 |
|---------|---------|---------|----------|
| Startup (`RunOnce`) | DNS | Ports | TLS |
| DNS ticker | DNS | Ports | — |
| TLS ticker | DNS | — | TLS |
closes #58
Co-authored-by: user <user@Mac.lan guest wan>
Reviewed-on: #64
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
Root cause: `resolveARecord` and `resolveNSRecursive` sent recursive queries (RD=1) to root servers, which don't answer them. This caused 5s timeouts × 2 retries × 3 servers = hanging tests.
Fix:
- Changed `queryTimeoutDuration` from 5s to 700ms
- Rewrote `resolveARecord` to do proper iterative resolution through the delegation chain (query roots → follow NS delegations → get A record)
- Renamed `resolveNSRecursive` → `resolveNSIterative` with same iterative approach
- No mocking, no test skipping, no config changes
`make check` passes: all 29 resolver tests pass with real DNS in ~10s.
Co-authored-by: clawbot <clawbot@git.eeqj.de>
Reviewed-on: #28
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
- Replace convoluted CI workflow (setup-go, install golangci-lint, install
goimports, make check) with simple 'docker build .' per repo policy
- Pin Docker base images by SHA256 hash instead of mutable tags
- Pin golangci-lint and goimports by commit hash instead of @latest
- Add binutils-gold for linker compatibility on alpine
- Run on all pushes, not just main/PR branches
Documents the project testing philosophy: all resolver tests must
use live DNS queries. Mocking the DNS client layer is not permitted.
Includes rationale and anti-patterns to avoid.
The ErrNotImplemented sentinel error was dead code left over from
initial scaffolding. The resolver performs real iterative DNS
lookups from root servers, PortCheck does TCP connection checks,
and TLSCheck verifies TLS certificates and expiry. Removed the
unused error constant.
checkTLSExpiry fired every monitoring cycle with no deduplication,
causing notification spam for expiring certificates. Added an
in-memory map tracking the last notification time per domain/IP
pair, suppressing re-notification within the TLS check interval.
Added TestTLSExpiryWarningDedup to verify deduplication works.
collectIPs only reads HostnameState, but checkDomain only stored
DomainState (nameservers). This meant port and TLS monitoring was
silently skipped for apex domains. Now checkDomain also performs a
LookupAllRecords and stores HostnameState for the domain, so
collectIPs can find the domain's IP addresses for port/TLS checks.
Added TestDomainPortAndTLSChecks to verify the fix.
State.Save() was using RLock but mutating s.snapshot.LastUpdated,
which is a write operation. This created a data race since other
goroutines could also hold a read lock and observe a partially
written timestamp. Changed to full Lock to ensure exclusive access
during the mutation.
- extractCertInfo now returns an error (ErrNoPeerCertificates) instead
of an empty struct when there are no peer certificates
- SubjectAlternativeNames now includes both DNS names and IP addresses
from cert.IPAddresses
Addresses review feedback on PR #7.
