| From: | "Ed L(dot)" <pgsql(at)bluepolka(dot)net> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "Ed L(dot)" <pggeneral(at)bluepolka(dot)net>, Laurette Cisneros <laurette(at)nextbus(dot)com> |
| Cc: | pgsql-admin(at)postgresql(dot)org |
| Subject: | Re: server won't shutdown |
| Date: | 2003-02-19 17:54:23 |
| Message-ID: | 200302191054.23651.pgsql@bluepolka.net |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
On Monday February 17 2003 8:06, Tom Lane wrote:
>
> You can demonstrate the problem by doing 'LISTEN foo' in one psql
> and then 'NOTIFY foo' in another, then attempting a fast shutdown
> --- the first psql's backend will ignore you until the psql session
> does something. The same path can be followed without LISTEN if the
> first psql is simply left idle while the second one does a bunch of
> catalog-updating work (a few cycles of VACUUM ANALYZE usually
> suffice).
Thanks, Tom. Your diagnosis sounds consistent with our experience. I
was able to reproduce it with the latter approach of running a bunch
of VACUUM ANALYZE commands. However, I have been unable to
demonstrate this on 7.3.1 using the following sequence:
session #1 launches psql
session #2 launches psql
session #1 does "listen foo;"
session #2 does "notify foo;"
session #3 does a fast shutdown;
All backends immediately shutdown, no hanging.
Anyway, I'm applying the patch and will advise if any further issues
show up.
Ed
| From | Date | Subject | |
|---|---|---|---|
| Next Message | P G | 2003-02-19 18:19:53 | What is the datetime type equivalent from v7.0.2 to v7.3.1??? |
| Previous Message | Peter Childs | 2003-02-19 16:33:39 | Use Postgres to check my passwords |