Re: 503 Backend fetch failed errors

On 11/8/18 9:35 AM, Magnus Hagander wrote:
>
>
> On Thu, Nov 8, 2018 at 9:31 AM Stefan Kaltenbrunner
> <stefan@kaltenbrunner.cc> wrote:
>
>     On 11/8/18 9:19 AM, Magnus Hagander wrote:
>      >
>      >
>      > On Thu, Nov 8, 2018 at 9:01 AM Stefan Kaltenbrunner
>      > <stefan@kaltenbrunner.cc> wrote:
>      >
>      >     On 11/7/18 6:40 PM, Rémi Zara wrote:
>      >      > Hi,
>      >
>      >     Hi Rémi!
>      >
>      >      >
>      >      > I’m getting a lot of these errors with coypu (several per
>     day),
>      >     but not systematically.
>      >      > Is this a problem on my end, or is this on the sever end ?
>      >      >
>      >      > Query for: stage=OK&animal=coypu&ts=1541573277
>      >      > Target:
>      >
>     https://buildfarm.postgresql.org/cgi-bin/pgstatus.pl/53b137e7c765b781699bbe73e3aec7751a8c4ab7
>      >      > Status Line: 503 Backend fetch failed
>      >      > Web txn failed with status: 1
>      >      > Query for: stage=OK&animal=coypu&ts=1541575423
>      >      > Target:
>      >
>     https://buildfarm.postgresql.org/cgi-bin/pgstatus.pl/1
>      >      > Status Line: 503 Backend fetch failed
>      >      > Web txn failed with status: 1
>      >
>      >     given the error this is something that is created by the varnish
>      >     instance that is in front of the buildfarm. On a quick look I
>     could
>      >     immediately figure out what the problem is - but it looks
>     like you (or
>      >     somebody else) tried at least to click one of the links above
>     using
>      >     hist
>      >     desktop browser and got an error about a missing branch
>     specification ;)
>      >
>      >
>      >
>      > AFAICT:
>      >
>      > A quick look in the logs indicates that the buildfarm is responding:
>      > -   RespHeader     Status: 492 bad branch parameter
>      >
>      > However, 492 is not a valid http status code, so Varnish can't
>     handle it
>      > and thus returns 503 failure to the client.
>
>     I think that is not the actual error that Rémi is experiencing- the 492
>     case (which is indeed an invalid http error code) only happens when one
>     actually klicks the link in the mail above(which I guess some did and
>     you found in the logs) because the actual BF client will add a
>     parameter
>     to the "Target" URL.
>
>     The actual "errors" dont seem to show up in the lighttpd logs afaiks.
>
>
> Oh, sorry. I was checking the one called "target", I assumed that was
> the URL that failed.
>
> Assuming for the original ones the ts is part of the URL, none of that
> is still in the logs. Or are they post parameters? Do we know exactly
> which URL is actually failing, and when (exactly) this happened?

well - most of the parameters to each url are in the error report (f.e.
"Query for: stage=OK&animal=coypu&ts=1541575423") I dunno whether Rémi
knows which branch that was for? - that one also has a unix timestamp,
though I "think" that is the timestamp from when the build started on
the bf-client and not the ts when the request was made)

Afaiks the two requests are not at all in the lighly log so only varnish
might have seen them (though its unclear what error it got while
connecting to lighty)