Use buffer pool for NewId

[jycor]

No description provided.

[github-actions]

	Main	PR
covering_index_scan_postgres	1926.37/s	1972.22/s	+2.3%
groupby_scan_postgres	132.32/s	131.91/s	-0.4%
index_join_postgres	680.16/s	676.90/s	-0.5%
index_join_scan_postgres	867.53/s	858.86/s	-1.0%
index_scan_postgres	25.56/s	25.68/s	+0.4%
oltp_delete_insert_postgres	876.81/s	857.09/s	-2.3%
oltp_insert	746.95/s	729.67/s	-2.4%
oltp_point_select	3084.40/s	3069.87/s	-0.5%
oltp_read_only	3087.94/s	3088.17/s	0.0%
oltp_read_write	2353.09/s	2352.71/s	-0.1%
oltp_update_index	770.16/s	774.39/s	+0.5%
oltp_update_non_index	822.04/s	823.80/s	+0.2%
oltp_write_only	1822.65/s	1816.84/s	-0.4%
select_random_points	1962.55/s	1936.11/s	-1.4%
select_random_ranges	1132.32/s	1167.80/s	+3.1%
table_scan_postgres	24.52/s	24.24/s	-1.2%
types_delete_insert_postgres	820.78/s	819.83/s	-0.2%
types_table_scan_postgres	10.52/s	10.55/s	+0.2%

[itoqa]

Commit: f22a1a4: 5 test cases ran, 5 passed ✅.

Summary

This run exercises core database metadata behavior end to end, including object lifecycle operations, identity mapping consistency, restart stability, and concurrent session isolation under contention. It also covers boundary handling for maximum-length identifiers, and overall behavior remained stable with clean state management across these functional and edge-case paths.

Safe to merge — this run shows no regressions, no new failures, and no still-failing issues attributable to this PR, while validating both normal flows and higher-risk concurrency and boundary conditions.

Tests run by Ito

Result	Severity	Type	Description
✅	—	Buffer	Schema-scoped create and drop operations for tables, sequences, and functions completed successfully, with stable OID lookups across repeated catalog reads and clean object removal at the end.
✅	—	Buffer	Catalog and lookup paths stayed consistent for all created objects, with 7/7 cross-path OID identity checks returning MATCH and clean teardown after drop.
✅	—	Concurrent	Catalog OIDs for built-in types, namespaces, and relations remained identical across first and second startup using the same data directory, confirming stable seeding behavior in this run.
✅	—	Concurrent	Four parallel SQL sessions each created and queried 20 session-scoped tables (80 total) with zero duplicate OIDs and no cross-session payload contamination, matching the expected isolation behavior for pooled ID generation.
✅	—	Format	A schema name at exactly 255 bytes remained queryable and supported create/alter/insert/drop metadata operations without ID format regression.

Tip

Reply with @itoqa to send us feedback on this test run.

[github-actions]

	Main	PR
Total	42090	42090
Successful	18275	18275
Failures	23815	23815
Partial Successes1	5335	5335

	Main	PR
Successful	43.4189%	43.4189%
Failures	56.5811%	56.5811%

Footnotes

1. These are tests that we're marking as Successful, however they do not match the expected output in some way. This is due to small differences, such as different wording on the error messages, or the column names being incorrect while the data itself is correct. ↩