[lesnik512]

Docs only, no issue. When validate_schema() finds that outbox_pending_idx (changed in 0.15) or outbox_leased_idx (new in 0.17) is missing, its error names the release and links to the matching upgrade migration. Only the Alembic guide's upgrade sections described this. The schema-validation page, where people land with the error in hand, didn't mention it.

The page gains one section. It shows a real error message, copied from _drift_entry_to_error's output for the 0.17 case, and a table mapping each index to its upgrade section. It also repeats the warning against alembic revision --autogenerate for these indexes, because autogenerate builds them without CONCURRENTLY.

mkdocs build --strict passes.

[github-actions]

Benchmark gate

✅ gate passed

scenario	msg/s	delete/msg	WALrec/msg	WALB/msg	fpi	upd	del	dead_tup
consumer/w1/b10	1321	1.000	6.72	922	242	5000	5000	10000
consumer/w1/b100	1391	1.000	7.11	1154	243	5000	5000	10000
consumer/w2/b10	1569	1.000	6.73	924	242	5000	5000	10000
consumer/w2/b100	1803	1.000	7.11	1154	243	5000	5000	10000
consumer/w4/b10	1642	1.000	6.74	949	242	5000	5000	10000
consumer/w4/b100	1890	1.000	7.12	1154	243	5000	5000	10000
consumer/w1/b100/tfbs100	5655	0.010	6.17	1171	265	5000	5000	10000
consumer/w1/b100/interval1-10	1396	1.000	7.11	1156	244	5000	5000	10000
producer/w1/b100	2916	0.000	3.04	597	1	0	0	0
producer/w1/b100/ppt1	983	0.000	4.04	635	0	0	0	0

Gated (fails the build): delete_calls + tuple counters (upd/del/ins) + the producer's insert_calls, exact; select_calls and other_calls within +2; wal_records within a 10% band; fetch_calls within 2x. msg/s, WAL bytes and total calls are informational (timing/FPI noise).