Re: [Pgbuildfarm-members] Odd error on Baiji

I've been seeing the following error on Baiji/HEAD for a couple of
days. Other animals on the same machine, and in fact other branches on
Baiji are fine. Any ideas what's up?

10/06/2008 23:00:00.64 Starting HEAD run...
Tue Jun 10 23:00:00 2008: buildfarm run for baiji:HEAD starting
[23:00:00] checking out source ...
[23:00:13] checking if build run needed ...
[23:00:18] copying source to pgsql.2516 ...
3291 File(s) copied
[23:01:29] running configure ...
[23:01:29] running make ...
[23:16:02] running make check ...
[23:18:37] running make install ...
[23:19:33] setting up db cluster ...
[23:19:41] starting db ...
[23:19:42] running make installcheck ...
[23:22:00] restarting db ...
[23:22:12] running make PL installcheck ...
[23:22:50] restarting db ...
[23:23:04] running make contrib installcheck ...
[23:25:55] stopping db ...
[23:25:58] running make ecpg check ...
[23:29:14] OK
Query for: stage=OK&animal=baiji&ts=1213135200
Target: http://www.pgbuildfarm.org/cgi-bin/pgstatus.pl/ddd9cde85edc0f5307a83b42ee5cc54d3a8d4a5a
Status Line: 491 bad ts parameter - 1213135200 (Tue Jun 10 22:00:00
2008 GMT) is in the future.
Content:
bad ts parameter - 1213135200 (Tue Jun 10 22:00:00 2008 GMT) is in the future

Web txn failed with status: 1
10/06/2008 23:29:19.70 Finished.

--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com

Dave Page wrote:
> I've been seeing the following error on Baiji/HEAD for a couple of
> days. Other animals on the same machine, and in fact other branches on
> Baiji are fine. Any ideas what's up?
>
> 10/06/2008 23:00:00.64 Starting HEAD run...
> Tue Jun 10 23:00:00 2008: buildfarm run for baiji:HEAD starting
> [23:00:00] checking out source ...
> [23:00:13] checking if build run needed ...
> [23:00:18] copying source to pgsql.2516 ...
> 3291 File(s) copied
> [23:01:29] running configure ...
> [23:01:29] running make ...
> [23:16:02] running make check ...
> [23:18:37] running make install ...
> [23:19:33] setting up db cluster ...
> [23:19:41] starting db ...
> [23:19:42] running make installcheck ...
> [23:22:00] restarting db ...
> [23:22:12] running make PL installcheck ...
> [23:22:50] restarting db ...
> [23:23:04] running make contrib installcheck ...
> [23:25:55] stopping db ...
> [23:25:58] running make ecpg check ...
> [23:29:14] OK
> Query for: stage=OK&animal=baiji&ts=1213135200
> Target: http://www.pgbuildfarm.org/cgi-bin/pgstatus.pl/ddd9cde85edc0f5307a83b42ee5cc54d3a8d4a5a
> Status Line: 491 bad ts parameter - 1213135200 (Tue Jun 10 22:00:00
> 2008 GMT) is in the future.
> Content:
> bad ts parameter - 1213135200 (Tue Jun 10 22:00:00 2008 GMT) is in the future
>
> Web txn failed with status: 1
> 10/06/2008 23:29:19.70 Finished.
>
>

It looks like the clock on baiji is very badly adjusted. From what I can
tell it's about 32 minutes out of whack. Buildfarm members should
ideally be using NTP or something similar.

cheers

andrew

On Wed, Jun 11, 2008 at 2:27 PM, Andrew Dunstan <andrew(at)dunslane(dot)net> wrote:
>
>
> It looks like the clock on baiji is very badly adjusted. From what I can
> tell it's about 32 minutes out of whack. Buildfarm members should ideally be
> using NTP or something similar.

Oh, thats odd - it is using NTP, but was out (and has now been kicked
into sync again). Any idea why it wasn't affecting Baiji/8.3 or any of
the Vaquita builds?

--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com

Dave Page wrote:
> On Wed, Jun 11, 2008 at 2:27 PM, Andrew Dunstan <andrew(at)dunslane(dot)net> wrote:
>
>> It looks like the clock on baiji is very badly adjusted. From what I can
>> tell it's about 32 minutes out of whack. Buildfarm members should ideally be
>> using NTP or something similar.
>>
>
> Oh, thats odd - it is using NTP, but was out (and has now been kicked
> into sync again). Any idea why it wasn't affecting Baiji/8.3 or any of
> the Vaquita builds?
>

If the build runs longer than the time difference then it won't fail
this check. You were just on the boundary.

cheers

andrew