test(b2): pin core edges found by mutation testing by ffumero2003 · Pull Request #249 · Backblaze/blazer

ffumero2003 · 2026-10-30T18:52:11Z

Context

Part of the test-integrity audit (#97). Mutation testing (Gremlins) found single-line changes to b2 production code that no test catches. This PR covers those items from #204's ### b2 section (slice C3).

Changes

Test-only. New tests in b2/accountinfo_edges_test.go and b2/core_edges_test.go:

Cache refresh after object-lock updates (baseline.go updateFileRetention and updateFileLegalHold): after an update, Info is fetched again instead of served from the stale cache.
Session-file size caps (accountinfo.go, write and read): a session of exactly 1,048,576 bytes is accepted and 1,048,577 bytes is rejected, on both write and read.
Delete on a folder object (b2.go Object.Delete): returns the non-regular-file error.
memoryBuffer.Close twice (buffer.go): idempotent, no panic.
Zero-progress wake (progress.go): no duplicate callback.
Vacuous "no secret in file" check removed from TestAccountInfoFileRoundTripAndPermissions: AuthData has no key field, so it could never fail. TestAccountInfoFileResumeAndClearOn401 already covers it.

Not covered here: three items need a small test hook in production code, which a test-only change can't add. They're filed as #246:

the temp-file cleanup error (accountinfo.go:135);
file.Sync() before rename (accountinfo.go:146);
returning a buffer to the pool (buffer.go:148).

Verification

Each mutation was applied alone; the new test failed, then passed after revert:

Mutation	Failing output
Retention or legal-hold cache refresh: error check flipped, or refresh deleted (4 breaks)	`post-update marker="before" calls=2 (want 3)`
Write cap `>` → `>=` / guard deleted	`Set 1048576-byte session: ... exceeds` / `Set 1048577-byte session error = <nil>`
Read cap `>` → `>=` / guard deleted	`AuthData() for 1048576-byte session: ok=false` / `... 1048577-byte session: ok=true`
Folder-delete guard negated / deleted	`Delete(folder) error = <nil>, want non-regular-file rejection`
Buffer `Close` guard flipped / deleted	`Close retained the buffer` / nil-pointer panic on second `Close`
Zero-wake `n > 0` flipped / deleted	`progress callbacks = [0 0]` / `[1 1 1]`

With no B2 credentials set: go test -race -count=5 -shuffle=on ./b2/..., a 30-run -race check of the buffer and progress tests, go vet ./..., GOOS=windows go vet ./b2/... and make ci pass.

`make ci` (tail)

go build ./...
go vet ./...
go mod tidy -diff
go mod verify
all modules verified
0 issues.
go test -race -shuffle=on ./...
ok  	github.com/Backblaze/blazer/b2	16.517s
ok  	github.com/Backblaze/blazer/b2/sync	4.412s
ok  	github.com/Backblaze/blazer/base	7.827s
ok  	github.com/Backblaze/blazer/internal/b2types	3.555s
ok  	github.com/Backblaze/blazer/internal/b2wiretest	2.319s
ok  	github.com/Backblaze/blazer/internal/blog	3.337s
ok  	github.com/Backblaze/blazer/internal/retry	3.128s
ok  	github.com/Backblaze/blazer/x/b2sim	2.111s
ok  	github.com/Backblaze/blazer/x/consistent	2.967s
ok  	github.com/Backblaze/blazer/x/transport	4.032s
ok  	github.com/Backblaze/blazer/x/window	2.679s
CGO_ENABLED=0 GOOS=linux GOARCH=amd64 go build ./...
CGO_ENABLED=0 GOOS=linux GOARCH=386 go build ./...
CGO_ENABLED=0 GOOS=linux GOARCH=arm go build ./...
CGO_ENABLED=0 GOOS=darwin GOARCH=arm64 go build ./...
CGO_ENABLED=0 GOOS=windows GOARCH=amd64 go build ./...
... scripts/*_test.sh: all ok

Part of #204

david-santacruz-backblaze

Two small things before approval, please check:

intent/b2-edge-assertions/ has only plan.md, while sibling slices (e.g. b2-mapping-edges) have intent.md, spec.md and plan.md. Could you add the other two, or drop plan.md and move the mutation table into the PR description?
AGENTS.md asks for the tail of make ci in the PR. Could you paste it? CI is green, so this is just for the record.

…e-assertions

Part of #204

ffumero2003 · 2026-10-30T21:10:55Z

@david-santacruz-backblaze Thanks, both done:

Intent folder: removed (617e15b). The mutation table was already in the PR description, so intent/b2-edge-assertions/ is gone; test: tighten minor test assertions (checklist) #204 is a checklist, so the description is its home.
make ci tail: added to the description, after syncing with main (1a06382): 0 issues., the race tests across every package, the five cross-builds and the script checks all pass.

#247 removed objectForDeleteOptionTest from b2_test.go while #249's TestObjectDeleteRejectsFolderStatus still used it, so the b2 tests no longer compile on main. Use the same setup as TestObjectDeleteAllowsCopyStatus.

…ext canceled (#252) * test(b2): reproduce writes after copy cancellation Pin delayed source reads that resume after copyContext returns and Writer.ReadFrom has been cancelled and closed. Part of #229 * fix(b2): stop a cancelled copy from writing after copyContext returns A cancelled copy could return while its source Read was still blocked. When that Read resumed, the copy goroutine could write to a Writer that the caller had already closed, causing the slice-bounds panic seen in CI run 40925001975. Guard destination writes and mark the guard stopped before returning on cancellation. Waiting for the copy goroutine is not an option because the source Read may block forever. Preserve the closer path that closes the body and waits for its copy to finish. Closes #229 * test(b2): match the abandoned-copy error with errors.Is and document ReadFrom cancellation Address the panel's nits on #229: the regression test compared the error text, so wrapping errCopyAbandoned would break it while the guard still works; and ReadFrom's godoc did not say that a Read already in progress on a non-seekable source can outlive a cancelled call. * docs(b2): state exactly what ReadFrom guarantees after a cancelled copy The comment said nothing a pending Read returns is written, but the guard only stops writes once copyContext returns: a Write already in progress, with data read before the cancellation, can still complete. Say what the code guarantees: after ReadFrom returns, nothing more from r reaches the Writer. * test(b2): pin ReadFrom cancellation during blocked error cleanup Hold cancel_large_file or WithCancelOnError callbacks behind channels after producer and worker upload failures. ReadFrom must return on cancellation before those channels are released, and the callback must receive the cancellation request result (#229). * fix(b2): release the copy guard before upload error cleanup Keep the existing Writer error path and release the copy guard only on terminal write errors. No buffer writes follow that hook, and the stopped check still gates every later write, so no source bytes can reach the destination after a cancelled copy returns. Release the error-state mutex before cancel_large_file and errf as well, so a producer observing a worker failure does not wait behind cleanup. Close joins its existing worker wait group on error exits to preserve delivery of the cleanup result to WithCancelOnError callers. Reject a separate deferred error-recording and cleanup mechanism: reusing setErr preserves first-error ownership and callback semantics without adding parallel state. The public API and ReadFrom godoc remain unchanged. Closes #229 * test(b2): pin error mutex release and Close cleanup waiting Hold the producer inside buffer allocation until a worker enters blocked cancellation cleanup, then release it into getErr while the copy guard is still held. This makes moving cleanup back under emux fail deterministically. Gate the worker cancellation request separately and verify that its nonblocking error callback has completed when Close returns. Removing the deferred worker wait now fails without making the callback depend on Close. Keep the existing cancellation cases and production code unchanged. * docs(b2): document Writer.Close callback wait * test(b2): make Writer.Close callback proof deterministic * test(b2): pin Writer.ReadFrom upload error precedence * fix(b2): preserve upload errors from Writer.ReadFrom * test(b2): pin ReadFrom caller cancellation with recorded transport errors * fix(b2): preserve caller cancellation cause in Writer.ReadFrom * test(b2): pin caller custom causes in Writer.ReadFrom * fix(b2): distinguish Writer cancellation from caller causes * test(b2): start ReadFrom deadline after upload setup * docs(b2): clarify Writer ReadFrom error identity * test(b2): bound copy cancellation and Close cleanup waits * docs(b2): clarify cancellation and breaking Close wait contract * test(b2): acknowledge cancellation stub body close results * test(b2): inline the folder-delete test setup removed by #247 #247 removed objectForDeleteOptionTest from b2_test.go while #249's TestObjectDeleteRejectsFolderStatus still used it, so the b2 tests no longer compile on main. Use the same setup as TestObjectDeleteAllowsCopyStatus.

ffumero2003 added 3 commits October 30, 2026 12:02

test(b2): pin account-info session size boundaries

4a35538

test(b2): pin core object and progress edges

cf84239

test(b2): record core edge mutation proofs

787e213

ffumero2003 added this to the 0.10.0 milestone Oct 30, 2026

ffumero2003 added the skip-changelog No changelog entry needed; skips the changelog check label Oct 30, 2026

ffumero2003 requested review from goanpeca and sophiecarreras as code owners October 30, 2026 18:52

ffumero2003 self-assigned this Oct 30, 2026

ffumero2003 requested a review from david-santacruz-backblaze October 30, 2026 18:52

david-santacruz-backblaze suggested changes Oct 30, 2026

View reviewed changes

ffumero2003 added 2 commits October 30, 2026 15:09

Merge remote-tracking branch 'origin/main' into test-issue-204-b2-edg…

1a06382

…e-assertions

chore: move the C3 mutation table to the PR description

617e15b

Part of #204

ffumero2003 requested a review from david-santacruz-backblaze October 30, 2026 21:10

david-santacruz-backblaze approved these changes Oct 30, 2026

View reviewed changes

ffumero2003 mentioned this pull request Oct 30, 2026

test(b2): pin writer retry caps, backoff delays and fake-only bucket tests #247

Merged

ffumero2003 merged commit 15b1d48 into main Oct 30, 2026
23 of 24 checks passed

ffumero2003 deleted the test-issue-204-b2-edge-assertions branch October 30, 2026 21:24

This was referenced Oct 30, 2026

fix(b2): return the upload error from Writer.ReadFrom instead of context canceled #252

Merged

test(b2): pin writer, reader and progress edges #257

Open

Navigation Menu

Repository navigation

test(b2): pin core edges found by mutation testing - #249

test(b2): pin core edges found by mutation testing#249
ffumero2003 merged 5 commits into
mainfrom
test-issue-204-b2-edge-assertions

ffumero2003 commented Oct 30, 2026 •

edited

Loading

Uh oh!

david-santacruz-backblaze left a comment

Uh oh!

ffumero2003 commented Oct 30, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

ffumero2003 commented Oct 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Context

Changes

Verification

make ci (tail)

Uh oh!

david-santacruz-backblaze left a comment

Choose a reason for hiding this comment

Uh oh!

ffumero2003 commented Oct 30, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

ffumero2003 commented Oct 30, 2026 •

edited

Loading

`make ci` (tail)