Fix data corruption on CP persistence inconsistency#452
Draft
koujl wants to merge 2 commits into
Draft
Conversation
During BR the IndexSvc journal and the AppendBlkAllocator CP superblock are flushed in parallel with no atomicity guarantee. If the process crashes after the index entry is durable but before the allocator superblock is written, the receiver resumes with the blk_id visible in the index but the allocator watermarks reset. The dedup skip path then bypasses commit_blk entirely, leaving the blocks silently available for reallocation by subsequent blob writes and corrupting the data on disk. The fix re-commits the blk_id recovered from the index entry so the allocator watermarks are restored before new allocations proceed.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
https://jirap.corp.ebay.com/browse/SDSTOR-23023
During BR the IndexSvc journal and the AppendBlkAllocator CP superblock are flushed in parallel with no atomicity guarantee. If the process crashes after the index entry is durable but before the allocator superblock is written, the receiver resumes with the blk_id visible in the index but the allocator watermarks reset. The dedup skip path then bypasses commit_blk entirely, leaving the blocks silently available for reallocation by subsequent blob writes and corrupting the data on disk.
The fix re-commits the blk_id recovered from the index entry so the allocator watermarks are restored before new allocations proceed.
Related PR: eBay/HomeStore#896
Also fixes a bug from #435 (https://jirap.corp.ebay.com/browse/SDSTOR-23162)