]> granicus.if.org Git - postgresql/commit
pg_upgrade: start/stop new server after pg_resetwal
authorBruce Momjian <bruce@momjian.us>
Tue, 20 Jun 2017 17:20:02 +0000 (13:20 -0400)
committerBruce Momjian <bruce@momjian.us>
Tue, 20 Jun 2017 17:20:10 +0000 (13:20 -0400)
commitb710248dd3f90c46bd4208e6bf1290048b9d76cd
treebb8e23e5b82f34205611cd8e1a33be28655da4c5
parenta69dfe5f40f77d586e8d4d9ecfc39d81612c7dda
pg_upgrade:  start/stop new server after pg_resetwal

When commit 0f33a719fdbb5d8c43839ea0d2c90cd03e2af2d2 removed the
instructions to start/stop the new cluster before running rsync, it was
now possible for pg_resetwal/pg_resetxlog to leave the final WAL record
at wal_level=minimum, preventing upgraded standby servers from
reconnecting.

This patch fixes that by having pg_upgrade unconditionally start/stop
the new cluster after pg_resetwal/pg_resetxlog has run.

Backpatch through 9.2 since, though the instructions were added in PG
9.5, they worked all the way back to 9.2.

Discussion: https://postgr.es/m/20170620171844.GC24975@momjian.us

Backpatch-through: 9.2
src/bin/pg_upgrade/check.c
src/bin/pg_upgrade/pg_upgrade.c
src/bin/pg_upgrade/pg_upgrade.h