Repository navigation
revert: zero every new block again (reverts #160) - #162
Merged
Merged
Conversation
The 48h soak on daa33a6 found another file's block with its first 176128 bytes zeroed on disk after a write into a hole on a different node. That range matches the head zero of the hole write, which points at #160's narrower zeroing. #145's full-block zeroing ran clean for 26,792 soak ops. Restore it until #160's cause is found. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This was referenced Nov 17, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Reverts #160 (daa33a6) while its data-corruption cause is found.
Why: the 48h soak on main daa33a6 failed after 5 minutes and 1,311 operations, with no fault injected. A read of /soak/d2/l1-f6 on node 1 returned the right size and wrong content, and the shared disk itself is wrong. Bytes [0, 176128) of that file's block 1 (physical block 61) are zeros, although operation 853 had written data there. That range is exactly the head zero of operation 1098, which ran just before on node 3: a write into a hole of a different file (/l2-c45) at block offset 172325. That write's own new block (physical 60) got its data and zeros correctly. Physical 61 got only the head zero. This is one file's data destroyed by another file's write. Details are in soak-failure-2-zeroed-block-head.md (vm-harness notes).
Evidence for reverting #160 specifically: the same soak driver ran clean on d7f5420 (#145's full-block zeroing, 26,792 ops) and on d85bbe0 (13,609 ops). The commits since are #159, #155 and #160, and #160 is the only one that changes zeroing. The cause isn't proven yet, so the soak reruns on this revert to confirm.
Cost: new data is written twice again (about 2×), as with #145. Correctness first. #160's narrower rule comes back once the cause is found and covered by a test.
This removes #160's own stale-bytes tests. #145's stale-bytes and cross-file pattern tests stay and still pass with the full-block zeroing.
🤖 Generated with Claude Code