Files

check / check (pull_request) Successful in 3m11s

Details

Isolate the lint cache per worktree and context-gate the native path (closes #99 )

Closes #80.

script/lint decided whether it could skip the pinned image by asking what
version was on PATH rather than where it was running, and cache isolation
is part of that same question. Both issues are that one defect.

The cache was one directory per repo, shared by every worktree on the
host. Two checkouts of this repo have identical Go file contents, so
their cache keys collide and golangci-lint replays the stored analysis,
including the paths recorded when it was produced. The loud direction of
that failure - a clean tree failed by a dirty sibling - is the harmless
one. The silent direction, a dirty tree passed by a clean sibling, is
another way for a gate here to report a green it did not earn.

The cache is now keyed on a digest of the worktree path, so a collision
is not possible, and it stays persistent per worktree: a warm run is
still seconds. Each cache records the worktree it belongs to and is
collected when that worktree is gone, so throwaway worktrees do not
accumulate caches; the tree lives under XDG_CACHE_HOME and is disposable
by definition. Collection makes the tree writable first, since the Go
module cache inside it is deliberately read-only, and never lets a
failure to tidy up fail the lint: the first cut did exactly that, and a
gate failing for a housekeeping reason is its own bug.

script/lint-audit is the backstop, and runs on every lint: it rejects
output citing any file that is not in the tree being linted, so a result
built out of another checkout's analysis is a hard error instead of a
silent pass. It runs on clean output too, because that is the case
nobody investigates. It never certifies that a run passed - it does not
look at whether there were findings - so it cannot itself become a gate
that reports a green.

golangci-lint's "parallel golangci-lint is running" refusal is now a
bounded retry rather than a verdict. It is not a lint result, and
exiting non-zero on it is indistinguishable to a caller from real
findings; issue #88 measured that a private cache does not remove the
contention. Exhausting the retries fails with a message that says the
tree was never analysed.

The native path now requires VAULTIK_LINT_IN_CONTAINER=1, which only the
Dockerfile's lint stage sets, in addition to matching the pin. A
developer's locally installed 2.12.2 is a different build reached by a
different code path and no longer bypasses the digest pin. /.dockerenv
was considered and rejected as the signal: dockerd creates it for
`docker run`, but it is not reliably present during a BuildKit
`docker build`, which is exactly the case the exception exists for.
Inside the container a version mismatch is now a hard error rather than
a fall-through, since there is no daemon there to fall through to.
Version detection uses `golangci-lint version --short`, the interface
meant for it, keeping the banner scrape only as a fallback.

script/bootstrap no longer prints "bootstrap complete" on a machine that
cannot run the gate. Docker missing, or present with an unreachable
daemon, is a hard failure naming exactly what breaks. Installing docker
from bootstrap was rejected: it needs root, a daemon, and on macOS a GUI
cask, so the attempt would itself fail in the common case and trade one
false success for a second failure mode.

TODO.md's claim that `make check` became "as trustworthy as
script/cibuild" is corrected to what README.md already said: only the
lint leg is equivalent, while tests and gofmt still run against the host
toolchain. README.md's requirements section gains docker and sqlite3.

Verified by reproduction, not inspection: two concurrent lints from two
worktrees of differing cleanliness each reported only their own findings
with no lock error; a real run made to report paths outside its tree
exits 1; a matching linter shimmed onto PATH is never invoked while the
pinned image runs; a PATH without docker makes bootstrap fail. script/
cibuild exits 0 with the lint layer executing in the pinned image, which
is what proves the in-container path still works.

2026-08-09 14:51:44 +00:00

18 KiB

Raw Blame History

Workflow

branch (from main)
do the work in Next Step
move Next Step to the top of Completed Steps
move the top item of Future Steps into Next Step
commit (TODO.md changes in the same commit as the work)
merge to main if the branch is not protected, otherwise open a PR
push

Status

pre-1.0

Next Step

Define remaining scope for a first tagged release and cut v0.1.0.

Completed Steps

2026-08-09: Isolated the lint cache per worktree and context-gated the native lint path (issues #99, #80). One defect seen twice: script/lint decided whether it could skip the pinned image by asking what version was on PATH rather than where it was running, and cache isolation is part of that same question. The cache was one directory per repo, shared by every worktree on the host, so two checkouts with identical Go file contents collided and golangci-lint replayed the stored analysis — paths and all. The loud direction of that failure (a clean tree failed by a dirty sibling) is the harmless one; the silent direction, a dirty tree passed by a clean sibling, is a sixth way for a gate here to report a green it did not earn. The cache is now keyed on a digest of the worktree path, and every run is audited by the new script/lint-audit, which rejects output citing any file that is not in the tree being linted — a backstop that runs on clean output too, because that is the case nobody investigates. Caches record the worktree they belong to and are collected when it disappears, so throwaway worktrees do not accumulate them; the whole tree lives under XDG_CACHE_HOME and is disposable. The parallel golangci-lint is running refusal is now a bounded retry rather than a verdict: it is not a lint result, and exiting non-zero on it is indistinguishable to a caller from real findings (#88 showed a private cache does not remove that contention). The native path now requires VAULTIK_LINT_IN_CONTAINER=1, set only by the Dockerfile lint stage, in addition to matching the pin, so a developer's locally installed 2.12.2 no longer bypasses the digest pin; /.dockerenv was rejected as the signal because dockerd creates it for docker run and it is not reliably present during a BuildKit docker build, which is the case the exception exists for. Version detection uses golangci-lint version --short with the old banner scrape kept only as a fallback. script/bootstrap no longer prints bootstrap complete on a machine that cannot run the gate: a missing docker, or one whose daemon is unreachable, is a hard failure naming exactly what breaks. Verification was by reproduction rather than inspection — two concurrent lints from two worktrees of differing cleanliness, a real run made to report an outside path, a matching linter shimmed onto PATH, and a PATH with docker removed — and is recorded on the pull request.
2026-08-09: Closed the fifth false-green mechanism (issues #93, #69). script/test omitted -count=1, so Go's test result cache could satisfy the gate outright: a second back-to-back make test printed the full set of 14 ok lines, every one marked (cached), having executed no test at all. Since ok <pkg> (cached) is an ok line, the "14 ok lines means the suite ran" signal this repo leans on was forgeable, one level below the Docker layer cache that #85 addressed. Fixed with -count=1 unconditionally rather than only in the container, because the pre-commit hook runs the same script and a gate honest only in CI is dishonest where people rely on it most; test-coverage got the same flag, and script/check inherits it by calling script/test. In the same area, make test-integration was deleted rather than made real: no file in the repo carried a build tag, so -tags=integration selected nothing and the target was an exact duplicate of make test. Tagging a subset was rejected because the entire suite runs in well under a minute, and a scheme whose failure mode is "some tests silently stopped running" is a poor trade for those seconds in a repo with this particular history. The -timeout was raised from 30s after measuring rather than after assuming: the standing claim that cold-cache compilation is charged against -timeout is false, disproved by a containerised run that spent 46s compiling and still reported per-package durations within noise of a warm host run. -timeout reaches the test binary as -test.timeout and its clock starts inside testing.M.Run, after the build. The real exposure was margin, not compilation. The 120s landed on is a deliberate, documented divergence from REPO_POLICIES.md:192, which mandates 30s, and from that file's canonical recipe at :212-214; the divergence is recorded in script/test's comment because REPO_POLICIES.md is org-canonical and not editable here, and issue #101 proposes amending the policy text upstream. Numbers and the full verification are recorded once, on the pull request, and are deliberately not restated here.
2026-08-09: Triaged all fifteen stale remote branches (issue #71) and deleted fourteen of them; the full per-branch disposition with evidence is recorded on that issue. Method mattered more than the outcome here: a three-dot git diff main...branch diffs from the merge base, so it replays everything that landed on main after the branch diverged and makes any old branch look like it holds unlanded work. That artifact is what made golangci-v2.12.2 appear to carry 126 files of unpushed changes when its tree was byte-identical to main's. Every containment claim here therefore rests on two-dot tip diffs, tree-hash equality, git cherry, and git branch -r --merged. Nine branches were plain ancestors of main with zero git cherry + commits. golangci-v2.12.2 had landed squashed as cc58583, whose tree hash equals the branch tip's exactly; note the hash recorded in the issue had gone stale because main advanced, so the check had to be redone rather than repeated. fix/sync-snapshot-cleanup was redundant, its one line already on main in syncWithRemote. feature/restore-progress-bar was superseded by printRestoreProgress and the disk-backed blob cache, and had become actively regressive — it would have deleted internal/blobgen/compress_test.go, the #28 regression test that landed separately. The two branches this issue was filed for both turned out to be closed questions that main had already moved past by a recorded decision, so neither was landed and no regression test was owed: ctime no longer exists anywhere in the codebase after 1c72a37 removed the column, the File.CTime field and every use (#54/#55), and change detection compares size, mtime, mode, uid and gid only, exactly as ARCHITECTURE.md documents — so the silently-skipped-file data-loss risk that made this a 1.0 item does not exist. The SQL allow-list branch would have reverted bfd7334, which replaced that very allow-list with regex sanitisation on review feedback, and would have broken getTableCount("snapshots") because its allow-list omits that table. feature/daemon-mode is untouched and deferred to #94 pending an owner decision, so it is the one branch besides main still on the remote. The stale TODO.md entry named in the issue needed no fix: e496aa3 had already removed it. No product code changed.
2026-08-09: Adopted the remaining upstream CHECK_EPOCH hardening (issue #91), closing the gap #85 knowingly left open. Four changes, all four decided as adopt upstream in sneak/prompts #26. (1) Each check stage now asserts [ -n "$CHECK_EPOCH" ] || exit 1 before running anything, so a build that supplies no --build-arg fails instead of lying. This is the item that mattered: an unset ARG is an empty string and an empty string is a stable cache key, so the second and every later bare docker build . on an unchanged tree replayed all three check layers and still exited 0 — and docker build . is the command REPO_POLICIES.md names verbatim as a thing that must be green, so the documented command was precisely the one that lied. Failed steps are never cached, which is what makes the guard fire on every invocation rather than once. (2) The epoch is now expanded into each check command rather than left as a bare declaration, so the cache miss no longer depends on BuildKit's unreferenced-ARG handling staying as it is, and the value appears in the build log. (3) script/cibuild uses epoch="$(date +%s%N)$$", unique per invocation rather than per second; %N alone is insufficient because busybox drops it silently and exits 0, and $$ is what makes the guarantee hold regardless. The bare-assignment form is kept deliberately — inlined in an argument, a failing substitution does not abort under set -eu and would yield an empty constant epoch, restoring the exact false green being fixed. (4) script/docker passes the same fresh arg, so the two entrypoints cannot disagree about whether the tree is green; local builds are almost always warm, which made it the likelier fooling in practice. The ARG placement from #85 is unchanged, below apk add, COPY go.mod go.sum and go mod download, so dependency layers still cache and the build is not cold. Verified by negative control rather than inspection — a bare docker build . run twice back to back, plus back-to-back pairs of both scripts and a host-side make check; the measurements are recorded once, in the PR verification comment, rather than restated here. .golangci.yml, the lint-stage FROM line and its digest, script/lint, REPO_POLICIES.md and .gitea/workflows/check.yml are all untouched.
2026-08-09: Stopped script/cibuild from reporting a green it did not earn (issue #85). A bare docker build . let Docker serve the check layers from the layer cache whenever the tree had not changed: the checks never executed and the build still exited 0. The fix is an ARG CHECK_EPOCH declared immediately above the check RUNs in both the lint stage and the builder stage (ARG scope is per-stage, so each declares its own), with script/cibuild assigning epoch="$(date +%s)" and passing --build-arg CHECK_EPOCH="$epoch". The assignment is separate on purpose: under set -eu a command substitution that fails inside an argument does not abort the script, which would leave an empty constant CHECK_EPOCH and restore the very false green being fixed. Placement is the rest of the point — the ARG sits below the apk add, COPY go.mod go.sum, and go mod download layers, so only the checks are invalidated and the dependency layers still cache. The guarantee is conditional on a fresh value rather than absolute: a bare docker build . gets an empty CHECK_EPOCH and can still serve the check layers from cache, which README.md and the Dockerfile now say plainly, with issue #91 tracking the upstream hardening (expanded ARG form, unset guard, per-invocation epoch, script/docker) that would close it. Verified by re-running the reproduction plus the withheld---build-arg counterfactual; the measurements are recorded once, in the PR #89 verification comment, rather than restated here. .golangci.yml, the lint-stage FROM line and its digest, script/lint, and .gitea/workflows/check.yml are all untouched.
2026-08-09: Corrected the Vaultik.UI doc comment (issue #84). It claimed the cli layer replaces the writer with a discarding one in --cron mode; the actual mechanism is UI.SetQuiet(true) in setupGlobals, which drops Begin/Complete/Info/Notice/Detail/ Progress/Banner but still emits Warning and Error. The --cron line in README.md said "Silent unless error", which understated what survives, and now names warnings too. The other --cron comments (internal/log/log.go, internal/cli/snapshot.go, internal/vaultik/snapshot.go) were audited and already accurate. Comments and docs only, no behavior change.
2026-08-09: Made snapshot list list the destination store without the private key (issue #64). The listing is now the union of the local index and a single streamed listing of the metadata/ prefix, with no age_secret_key gate — the manifest is unencrypted, so a host holding only the public key can enumerate its own backups and a host that lost its local index can still see them. A remote-only snapshot's hostname and name are deliberately not recovered (they are not recoverable without the private key, and making them so would undo the privacy property tracked in issue #81); such rows are labelled by an abbreviation of their remote key and carry the real timestamp and compressed size from the manifest, with <remote only> in the two columns that require the local index. Local-only snapshots are reported as drift, and the hint now names vaultik prune, which exists, instead of vaultik snapshot cleanup, which does not. reportRemoteDrift collapsed into the merged view. Every remote manifest read in the codebase now goes through downloadManifestByKey, so issue #81 has one call site to change. Review rework: snapshot timestamps now normalize to UTC in scanSnapshotRows, the one place they enter the domain, so the merged TIMESTAMP column cannot show local time for a locally tracked row and UTC for a remote-only row on a non-UTC host; GetIncompleteByHostname was folded onto that same scanner. --json now reports the unreadable-manifest count and the 1000-row truncation on stderr instead of returning a silently short document (the document's shape is unchanged). The two per-snapshot log.Warn calls on the listing path now route through the same JSON-aware writer as the existing workaround, so one corrupt manifest can no longer put a log line on stdout ahead of the document and break | jq — still a local workaround pending issue #82. Verified with script/cibuild and with an uncached make check (0 issues., no cached test packages), plus end to end against a file:// destination with no secret key present.
2026-08-09: Closed the gap between make lint and CI (issue #78). script/lint now runs the digest-pinned golangci-lint image taken from the Dockerfile lint stage, which is the single source of truth for the linter version; the duplicate pin in the Makefile deps target and the unpinned golangci-lint install in script/bootstrap are gone. A golangci-lint on PATH is used only when its version is exactly the pinned one (which is how the lint stage runs it inside the container); anything else goes through Docker, and a missing or unreachable Docker daemon is a hard error rather than a silent fallback. Only the lint leg of make check became equivalent to script/cibuild; its tests and gofmt still run on the host against the host toolchain, as README.md states. An earlier version of this entry claimed make check was "as trustworthy as script/cibuild" outright, which overstated it; corrected under issue #80.
2026-08-09: Finished the lint remediation under the canonical .golangci.yml (issue #61, which also unblocks issue #59). The remaining findings were fixed behavior-preservingly: wsl_v5 whitespace, sqlclosecheck, and prealloc. The sqlclosecheck sites now close sql.Rows in a deferred closure instead of via the CloseRows helper, which the linter could not see through. Only the revive package-name findings remain suppressed, with per-site //nolint directives; the package-rename question behind them is tracked in issue #76. Verified with script/cibuild, which exits 0 — that is the only trustworthy gate, because script/lint runs whatever golangci-lint happens to be on PATH rather than the pinned v2.12.2 that CI and the Dockerfile use, so make check can report green on findings CI still fails. That tooling gap is tracked in issue #78.
2026-08-09: The earlier next step "reconcile the uncommitted ARCHITECTURE.md edits on main" needed no work: the working tree is clean and ARCHITECTURE.md is committed on main.
2026-08-07: Updated golangci-lint to v2.12.2 everywhere it is pinned (Dockerfile lint stage, Makefile deps target), replaced .golangci.yml with the canonical config (v2 schema, default: all), and remediated the bulk of the lint findings it surfaced (issue #61): behavior-preserving fixes across every package, 2,990 findings down to 80. make test and make fmt-check were green at that point but make lint was still red; the commit message claiming make check was green was wrong.
2026-08-07: Added the standard .golangci.yml and .editorconfig (issue #59); lint findings under the new config are tracked in issue #61. script/bootstrap now installs sqlite3 (needed by tests).
2026-07-07 Adopted scripts-to-rule-them-all: script/ entrypoints, Makefile shims, README Entrypoints section
2026-07-02: Consolidated CLI verbs, retired overlapping commands; bound the local index to its backup destination URL.
2026-06-28: snapshot rm now removes metadata only and prints the prune command; restore skips chown when running as non-root.
2026-06-26: Snapshot IDs hashed at the storage boundary; snapshot list made resilient to bad remote entries.
2026-06-24: Collapsed snapshot prune into vaultik prune; restore streams blobs to disk and restores files in blob-locality order; cron output fixes.
2026-06-17: Restore overhaul: ReadAt chunk reads from cached blobs, reference-counted blob sweeper, integration tests; new internal/ui output layer, banner, and progress lines.
2025-12-18: Added ARCHITECTURE.md and godoc coverage for exported API.
2025-07-26: End-to-end integration tests; manifest format refactor; renamed backup to snapshot; afero filesystem abstraction.
2025-07-20: Initial design and implementation: cobra + fx CLI skeleton, SQLite index database, UUID blob storage with streaming chunking.

Future Steps

None queued; the release-scoping item is now the Next Step.