From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Jeff Boes <jboes(at)nexcerpt(dot)com> |
Cc: | pgsql-admin(at)postgresql(dot)org |
Subject: | Re: Contents of pg_listener linger? |
Date: | 2004-02-06 16:26:01 |
Message-ID: | 711.1076084761@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Jeff Boes <jboes(at)nexcerpt(dot)com> writes:
> Yup, that seems to be the case. When I "NOTIFY" these, they disappear.
> Any other way to clean them up?
If you're superuser you can delete rows all you want. You could
probably cons up something that deletes rows with PIDs not found in
the pg_stat_activity view.
I'm more interested in how the rows got there though. The bugs I was
thinking of were surely stomped long before 7.3.4, so if you're not
suffering crashes then you must have found a new one. Please keep an
eye on it and see if you can figure out the circumstances in which a row
gets left behind. Since you know the PIDs of the delinquent backends,
maybe you could look at the last related entries in the postmaster log
(turn on log_pid if you're not using syslog for logging).
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | hal | 2004-02-06 16:54:42 | HELP: optimizing shared mem |
Previous Message | Jeff Boes | 2004-02-06 16:18:23 | Re: Contents of pg_listener linger? |