fix(replica): fsync directory after durable renames#1344
Open
corylanou wants to merge 1 commit into
Open
Conversation
Add internal.FsyncDir and call it after every rename that finalizes a durable file: both restore output paths, local L0 LTX staging, and the file replica backend's LTX write. Refactor WriteTXIDFile's inline directory sync onto the helper. Without the directory fsync a crash shortly after a successful rename can lose the directory entry.
PR Build Metrics✅ All clear — no issues detected
Binary Size
Dependency ChangesNo dependency changes. govulncheck OutputBuild Info
🤖 Updated on each push. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Adds
internal.FsyncDirand calls it after every rename that finalizes a durable file:Replica.Restoreoutput paths (main + V3),db.go,filereplica backend'sWriteLTXFile.WriteTXIDFile's existing inline directory sync is refactored onto the same helper, so the crash-durability pattern (temp + fsync + rename + dir-sync) now lives in one place.Motivation and Context
Without a directory fsync, a crash shortly after a successful
os.Renamecan lose the directory entry — a restore that reported success can leave no database file, and the file backend can lose a replica LTX object it reported as written.WriteTXIDFilealready did this correctly; the other rename sites did not. The two non-restore sites were found by review while fixing the restore path (same defect class).Fixes #1342
How Has This Been Tested?
go build ./...,go vet ./..., andgo test -racefor the root,file, andinternalpackages all pass. Durability-on-crash itself is not directly testable without fault injection; the change is the establishedWriteTXIDFilepattern applied to the remaining sites.Types of changes
Checklist
go fmt,go vet)go test ./...)