Michael van Rooyen <michael@loot.co.za> writes:
> I'm trying to get to the bottom of a performance issue on a server
> running PostgreSQL 9.3.1 on Centos 5.
Hm ... it seems pretty suspicious that all of these examples take just
about exactly 1 second longer than you might expect. I'm wondering
if there is something sitting on an exclusive table lock somewhere,
and releasing it after 1 second.
In particular, this looks quite a bit like the old behavior of autovacuum
when it was trying to truncate empty pages off the end of a relation ---
it would hold off other accesses to the table until deadlock_timeout
elapsed, whereupon it'd get kicked off the exclusive lock (and have to
retry the truncation next time). Are you *sure* this server is running
9.3.1, and not something pre-9.3?
regards, tom lane