Re: 8.3.0 Core with concurrent vacuum fulls

On Wed, Mar 5, 2008 at 10:31 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> "Gavin M. Roy" <gmr(at)myyearbook(dot)com> writes:
> > 2008-03-04 05:45:47 EST [6698]: [1-1] LOG: process 6698 still waiting
> for
> > AccessShareLock on relation 1247 of database 16385 after 1001.519 ms
> > 2008-03-04 05:45:47 EST [6698]: [2-1] STATEMENT: VACUUM FULL
> > autograph.autograph_creators
> > 2008-03-04 05:46:28 EST [6730]: [1-1] LOG: process 6730 still waiting
> for
> > AccessShareLock on relation 1247 of database 16385 after 1000.887 ms
> > 2008-03-04 05:46:28 EST [6730]: [2-1] STATEMENT: VACUUM FULL
> > lunchmoney.totals
> > 2008-03-04 05:47:55 EST [3809]: [18-1] LOG: server process (PID 6742)
> was
> > terminated by signal 6: Aborted
> > 2008-03-04 05:47:55 EST [3809]: [19-1] LOG: terminating any other
> active
> > server processes
> > 2008-03-04 05:47:55 EST [6741]: [12-1] WARNING: terminating connection
> > because of crash of another server process
>
> How annoying ... the PANIC message doesn't seem to have reached the log.
> elog.c is careful to fflush(stderr) before abort(), so that isn't
> supposed to happen. But it looks like you are using syslog for logging
> (correct?) so maybe this is a problem with the syslog implementation
> you're using. What's the platform exactly?
>
> I wonder if it'd be reasonable to put a closelog() call just before
> the abort() ...
>
> regards, tom lane
>

The panic may have made it if this is what you were looking for:

2008-03-04 05:45:20 EST [6742]: [7-1] PANIC: deadlock detected
2008-03-04 05:58:33 EST [8751]: [3-1] PANIC: deadlock detected

(it cored twice before I lowered the concurrency of vacuums)