Current issue state, recent activity, and per-issue timelines from the indexed issue data.
| Date | Opened | Closed | Comments | Events | Open Backlog |
|---|---|---|---|---|---|
| 2026-09-20 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-19 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-18 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-17 | 1 | 4 | 2 | 5 | 28 |
| 2026-09-16 | 4 | 0 | 0 | 0 | 0 |
| 2026-09-15 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-14 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-13 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-12 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-11 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-10 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-09 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-08 | 0 | 0 | 0 | 0 | 0 |
| 2026-09-07 | 0 | 0 | 0 | 0 | 0 |
Opened: 5
Closed: 4
Comments: 2
Events: 5
| Issue | Author | State | Labels | Comments | Reactions | Updated |
|---|---|---|---|---|---|---|
#315 Boolean queries (unranked AND/OR/NOT) Opened 5 months ago | tjgreen42 | closed - completed | enhancement scope:medium | 0 | 0 | 3 days ago |
#486 Global memtable cache byte accounting leaks across dropped objects Opened 17 days ago | tjgreen42 | closed - not_planned | bug | 3 | 0 | 4 days ago |
#376 ci/bench: MS MARCO concurrent validation regression — queries 9351 and 9340 fail per-rank score check on main (post-#366) Opened 4 months ago | tjgreen42 | closed - completed | bug | 2 | 0 | 4 days ago |
#321 refactor: eliminate TpAliveBitset struct, modify bitset pages in-place during VACUUM Opened 5 months ago | tjgreen42 | closed - completed | performance | 0 | 0 | 4 days ago |
#499 Optimize Boolean queries with cache-backed memtable evaluation Opened 4 days ago | tjgreen42 | open | No labels | 0 | 0 | 4 days ago |
#498 Reduce peak memory during large BM25 index builds Opened 4 days ago | tjgreen42 | open | scope:large performance | 0 | 0 | 4 days ago |
#497 Investigate poor read-only disjunction performance in PlanetScale TIN benchmark Opened 4 days ago | tjgreen42 | open | scope:large performance | 0 | 0 | 4 days ago |
#496 Benchmark Boolean, phrase, mixed, and COUNT workloads after query support lands Opened 4 days ago | tjgreen42 | open | enhancement scope:medium performance | 0 | 0 | 4 days ago |
#495 Prevent memtable spill starvation under continuous ranked queries Opened 4 days ago | tjgreen42 | open | bug scope:medium performance | 0 | 0 | 4 days ago |
#313 Background segment compaction Opened 5 months ago | tjgreen42 | open | enhancement scope:medium | 1 | 0 | 12 days ago |
#456 On-disk uint32 format limits cap a fully-merged segment at a 4 GiB string pool / 268M terms. Is segment splitting (or a wider format) the intended path for very large corpora? Opened 1 month ago | YaserJaradeh | open | enhancement | 0 | 0 | 14 days ago |
#485 Compaction entry points can act on stale per-index state after concurrent index DDL Opened 17 days ago | tjgreen42 | open | enhancement | 0 | 0 | 17 days ago |
#484 Confine at-capacity segments to the deepest reachable level Opened 17 days ago | tjgreen42 | open | enhancement scope:medium performance | 0 | 0 | 17 days ago |
#482 Hot-standby BM25 query can fail with invalid segment header after primary merge/reclaim Opened 18 days ago | me1t0ut | open | bug | 0 | 0 | 18 days ago |
#481 index does not exist Opened 19 days ago | panda-hub102 | closed - completed | bug | 0 | 0 | 18 days ago |
#473 Enforce segment format limits consistently across build and merge Opened 23 days ago | tjgreen42 | open | bug scope:medium | 0 | 0 | 23 days ago |
#470 REINDEX orphans memtable cache accounting, permanently leaking shared memory budget Opened 26 days ago | tjgreen42 | open | No labels | 0 | 0 | 26 days ago |
#466 Package Release: harden arm64 QEMU builds against transient failures Opened 27 days ago | tjgreen42 | closed - completed | No labels | 0 | 0 | 26 days ago |
#469 Tombstone chain walks don't bound the next-page pointer, so corruption bypasses self-heal Opened 27 days ago | tjgreen42 | open | No labels | 0 | 0 | 27 days ago |
#468 tp_record_free_index_page() reads a block with no bounds check Opened 27 days ago | tjgreen42 | open | No labels | 0 | 0 | 27 days ago |
#320 perf: VACUUM identify phase should avoid bulk CTID preload Opened 5 months ago | tjgreen42 | closed - completed | performance | 0 | 0 | 27 days ago |
#464 Shared index state is keyed on index OID alone, so template-cloned databases collide Opened 28 days ago | lmwilki | open | No labels | 0 | 0 | 28 days ago |
#433 Serial `CREATE INDEX USING bm25` dies with `arena: exceeded maximum capacity` because the flush budget is never clamped to the arena's addressable size Opened 2 months ago | YaserJaradeh | closed - completed | bug | 0 | 0 | 1 month ago |
#432 `uint32` handling of segment sections larger than 4 GiB: one silent write truncation and one silent read wrap Opened 2 months ago | YaserJaradeh | closed - completed | bug | 0 | 0 | 1 month ago |
#431 `invalid memory alloc request size` during parallel `CREATE INDEX USING bm25` on a large corpus Opened 2 months ago | YaserJaradeh | closed - completed | bug | 1 | 0 | 1 month ago |