Robert Creager <Robert_Creager(at)LogicalChaos(dot)org> writes:
> postgres 29304 20209 0 17:37 pts/4 00:00:00 postgres: robert tassiv 192.168.0.250 idle in transaction
Almost certainly, that guy is holding the lock REINDEX wants. The
VACUUM is probably backed up behind the REINDEX. If you want to
convince me it's a deadlock, you'll need to exhibit pg_locks contents.
regards, tom lane