Re: gitweb not working too well

From: Magnus Hagander <magnus(at)hagander(dot)net>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: PostgreSQL WWW <pgsql-www(at)lists(dot)postgresql(dot)org>
Subject: Re: gitweb not working too well
Date: 2018-05-23 15:38:31
Message-ID: CABUevEwF58iEfwXA6ockFd04ftLDhowzcKZT2tiiy8Rh-G6ctQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-www

On Wed, May 23, 2018 at 5:33 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Magnus Hagander <magnus(at)hagander(dot)net> writes:
> > On Wed, May 23, 2018 at 5:05 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> >> I've been getting a lot of Varnish "Error 503 Backend fetch failed"
> >> from https://git.postgresql.org/gitweb/ sub-pages yesterday and
> >> today. Is anybody looking into that?
>
> > Have you seen any pattern to when it's happening?
>
> Not particularly. When the problem occurs, it seems to persist for
> awhile (several minutes at least), but beyond that I can't think of
> anything.
>

I think this may simply be the result of overload. We currently limit the
system to two concurrent cache misses to gitweb, and while they are still
running everything else gets put in a backlog. We time them out fairly
aggressively, but given the performance of gitweb.

I'm going to give that VM some extra resources and restart it -- let's see
if that helps a bit.

--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>

In response to

Browse pgsql-www by date

  From Date Subject
Next Message Bruce Momjian 2018-05-23 22:41:16 Re: SCRAM with channel binding downgrade attack
Previous Message Tom Lane 2018-05-23 15:33:35 Re: gitweb not working too well