(clickhouse.com)

115 points saisrirampur | 1 comments | 15 Jul 25 04:43 UTC | HN request time: 0.206s | source

Show context

OkPin ◴[15 Jul 25 11:43 UTC] No.44570136[source]▶

Fascinating root cause: a missing CHECK_FOR_INTERRUPTS() left pg_create_logical_replication_slot basically unkillable on hot standbys. Simple fix, but huge impact.

Makes me wonder how many other Postgres processes might ignore SIGTERM under edge conditions. Do folks here test signal handling during failovers or replica maintenance? Seems like something worth adding to chaos tests.

replies(1): >>44577791 #

1. MuffinFlavored ◴[16 Jul 25 01:26 UTC] No.44577791[source]▶

>>44570136 #

Is that the root cause though? Why did this process get into a position that it needed to be manually killed/restarted, isn't that the real problem?

↑

When Sigterm Does Nothing: A Postgres Mystery