Repository navigation
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #689 +/- ##
==========================================
+ Coverage 91.53% 91.91% +0.37%
==========================================
Files 120 121 +1
Lines 23684 25054 +1370
==========================================
+ Hits 21680 23028 +1348
- Misses 2004 2026 +22
☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Changelog PreviewThis PR lands in these workspaces:
📋 Changelog PreviewThis is how your changes will appear in the changelog. New Features
🤖 This preview updates automatically when you update the PR. |
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 2 potential issues.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 19c2d94. Configure here.
| .write(batch) | ||
| .context(ErrorKind::BackendFailure, "writing rocksdb upload marker")?; | ||
| Ok(()) | ||
| } |
There was a problem hiding this comment.
Upload markers skip change stream
Medium Severity
create_upload_marker and delete_upload_marker never call the change stream, so marker lifetime is invisible to cost tracking. BigTableBackend issues begin_write/commit_write on create and commit_delete on a live delete, which this backend is supposed to match.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit 19c2d94. Configure here.


objectstore-servicenow implements the tiered-storage high-volume contract (Backend + HighVolumeBackend) over an embedded RocksDB, alongside the BigTable and in-memory implementations. Rows mirror Bigtable's cell model as flat KV cells keyed by storage path plus a two-byte column suffix, with exact point gets only.Every cell value is framed with an 8-byte deadline prefix, which drives a compaction filter that physically drops expired rows without decoding their payloads. A single write lock serializes all mutating read-modify-write sequences while reads use a snapshot, reproducing
InMemoryBackend's atomicity so compare-and-write cannot interleave. Change-stream write/update/delete reporting matchesBigTableBackend.