Once we start getting that error in the postgres log, there is no autovacuum entry. Only that error message once every minute (and as i wrote in my last email, the same message all the time).
The question is what is causing postgres to stop working. Autovacuum or another reason? If it's another reason ... what could it be?
Excerpts from Pablo Delgado Díaz-Pache's message of lun nov 15 04:52:53 -0300 2010:
> > But it strikes me that the code comment is wrong in one significant way: > > if the postmaster were failing to heed SIGUSR1 at all, you could reach > > the timeout here, because the fork-failed signal wouldn't get sent.
Hmm, yeah, I guess I wasn't considering the idea that postmaster was so busy with something else that it wouldn't be able to respond to the launcher's requests. As you say, if it went away entirely, autovacuum would also suicide.
> > Could you try strace'ing the postmaster process to see what it's doing > > when this is happening? > > I definitely will.
Yes, please.
-- Álvaro Herrera <alvherre@commandprompt.com> The PostgreSQL Company - Command Prompt, Inc. PostgreSQL Replication, Consulting, Custom Development, 24x7 support
