From: | "scott(dot)marlowe" <scott(dot)marlowe(at)ihs(dot)com> |
---|---|
To: | Jan Weerts <j(dot)weerts(at)i-views(dot)de> |
Cc: | "'Dorward Villaruz'" <dorwardv(at)ntsp(dot)nec(dot)co(dot)jp>, "'Postgres General'" <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: help in starting up / shutting down postgres as another |
Date: | 2002-11-22 16:37:30 |
Message-ID: | Pine.LNX.4.33.0211220935120.25374-100000@css120.ihs.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Fri, 22 Nov 2002, Jan Weerts wrote:
> > we have an application that needs to restart postgres if it
> >crashes.
>
> This is not a solution, postgres should provide, as it is OS-related.
> As Tom already stated, crashes are rare (we never had one in the last
> years) and if it crashed it obviously cannot control anything else :-)
>
> If crashes are not the main problem, I'd recommend sudo (exists for
> many unix derivates) or equivalents. It can selectively grant the
> "other" user the right to execute a postgres startup script/executable
> as the postgres user, without the need to share passwords or grant
> more rights than necessary.
I just wanna chime in on the "postgres never crashes" thread.
I haven't had postgres crash since I handed 6.5.3 an unconstrained join on
two rather large tables. And even that took almost an hour before
postgres used up all memory and swap and the linux kernel killed it, so
technically, postgres didn't crash, it just didn't know when to stop
trying to deliver an undeliverable dataset.
Postgres doesn't crash. Not on good hardware (i.e. error free memory and
CPU).
From | Date | Subject | |
---|---|---|---|
Next Message | Doug McNaught | 2002-11-22 16:38:49 | Re: Lack of use of indexes |
Previous Message | Tom Lane | 2002-11-22 16:34:13 | Re: improving a badly optimized query |