From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Jason Palmer" <jpalmer(at)dlife(dot)com> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: BUG #2710: Intermittent hangs on sequence generation |
Date: | 2006-10-21 03:53:24 |
Message-ID: | 21127.1161402804@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
"Jason Palmer" <jpalmer(at)dlife(dot)com> writes:
> ... What we experience is bizarre. We can always
> create tables with no sequences. But occasionally the database will hang
> when we attempt to create a table with a sequence or foreign key.
Yup, that's pretty bizarre. Is the backend consuming CPU (or I/O) in
this state, or is it just waiting? If the latter, it's presumably
blocked on some lock, and you need to find out what. First try looking
in pg_locks to see if it's blocked on a user-level lock. If you don't
see anything there, then it's blocked on some internal lock (LWLock or
spinlock) --- you'll need to attach to the backend process with gdb and
get a stack trace to get some clue what's up.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Eisentraut | 2006-10-21 10:34:59 | Re: BUG #2709: how to start postgresql at startup |
Previous Message | Tom Lane | 2006-10-20 22:17:11 | Re: Bug in 8.1.5: cannot add literal timestamp value and a interval. |