Improve some error messages with invalid indexes for REINDEX CONCURRENTLY
An invalid index is skipped when doing REINDEX CONCURRENTLY at table
level, with INDEX_CORRUPTED used as errcode. This is confusing,
because an invalid index could exist after an interruption. The errcode
is switched to OB
Fix issues in binary_upgrade_logical_slot_has_caught_up().
The commit 29d0a77fa6 labelled binary_upgrade_logical_slot_has_caught_up()
as a non-strict function to allow providing a better error message to callers
in case the passed slot_name is NULL. On further discussion, it seems that
it is not h
Fix assertion failure with REINDEX and event triggers
A REINDEX CONCURRENTLY run on a table with no indexes would always pop
the topmost snapshot from the active snapshot stack, making the snapshot
handling inconsistent between the multiple-relation and single-relation
cases. This commit slightly
Suppress -Wunused-result warning about write().
pg_test_fsync's signal_cleanup() intentionally ignores the write()
result since there's not much we could do about it, but certain
compilers make that harder than it ought to be.
This was missed in commit 52e98d4502.
Reviewed-by: Tristan Partin, Pe
Use signal-safe functions in signal handler
According to signal-safety(7), exit(3) and puts(3) are not safe to call
in a signal handler.
Author: Tristan Partin
Discussion:
https://www.postgresql.org/message-id/flat/CTVDKVZCCVSY.1XQ87UL50KQRD%40gonk
Branch
--
master
Details
---
https:/
