You are viewing a plain text version of this content. The canonical link for it is here.
Posted to builds@apache.org by Zoran Regvart <zo...@regvart.com> on 2020/11/02 21:21:25 UTC

Re: Hung website job

Seems that this has happened again with this job:

https://ci-builds.apache.org/job/Lucene/job/Solr-reference-guide-master/404/

as in the previous case there seems to be a snag when checking out from git.

Two cases with the same symptoms, perhaps it's worth investigating?

zoran

On Tue, Oct 27, 2020 at 2:51 PM Andrei Sekretenko <as...@d2iq.com> wrote:
>
> Hi Zoran,
> sure, this particular job can be cancelled/retried.
>
> From looking at the log, there seems to be some infrastructure issue on the ASF Jenkins / Git plugin/system/... level:
> the build has yet to reach the build script, and looks stuck at the git checkout by the SCM plugin.
> This might be worth investigating on the infra side (especially given that Jenkins hadn't aborted the stuck job in 300 minutes; I'm not 100% sure if Jenkins is supposed to abort a stuck job at the SCM checkout step, though).
>
> On Tue, Oct 27, 2020 at 1:32 PM Zoran Regvart <zo...@regvart.com> wrote:
>>
>> Hi Mesos folk,
>> perhaps someone from your side can take a look?
>>
>> thanks!
>>
>> zoran
>> ---------- Forwarded message ---------
>> From: Zoran Regvart <zo...@regvart.com>
>> Date: Tue, Oct 27, 2020 at 11:49 AM
>> Subject: Hung website job
>> To: builds <bu...@apache.org>
>>
>>
>> Hi Builders,
>> seems that a job has hung[1] on the website1 node preventing other
>> website jobs to schedule.
>>
>> Can it be canceled/retried?
>>
>> zoran
>> [1] https://ci-builds.apache.org/job/Mesos/job/Mesos-Websitebot/716/
>> --
>> Zoran Regvart
>>
>>
>> --
>> Zoran Regvart
>
>
>
> --
> --
> Andrei Sekretenko



-- 
Zoran Regvart

Re: Hung website job

Posted by Matt Sicker <bo...@gmail.com>.
That’s all configurable per job, etc.

Depending on how big your repo is, you might be able to use lighter weight
checkouts.

On Mon, Nov 2, 2020 at 16:01 Dave Fisher <wa...@apache.org> wrote:

> Hi -
>
> I’ve been sledgehammering gitbox with OpenOffice.org migrations. I’m
> taking a break as I appear to have hit a JBake limit.
>
> Jenkins has a 10 minute timeout git setup.
>
> Regards,
> Dave
>
> Sent from my iPhone
>
> > On Nov 2, 2020, at 1:28 PM, Chris Lambertus <cm...@apache.org> wrote:
> >
> > Try directing your git checkouts to github. There are some
> circumstances when gitbox becomes overloaded, let's see if the problem goes
> away with github.
> >
> >
> >
> >
> >
> >
> >> On Nov 2, 2020, at 1:21 PM, Zoran Regvart <zo...@regvart.com> wrote:
> >>
> >> Seems that this has happened again with this job:
> >>
> >>
> https://ci-builds.apache.org/job/Lucene/job/Solr-reference-guide-master/404/
> >>
> >> as in the previous case there seems to be a snag when checking out from
> git.
> >>
> >> Two cases with the same symptoms, perhaps it's worth investigating?
> >>
> >> zoran
> >>
> >>> On Tue, Oct 27, 2020 at 2:51 PM Andrei Sekretenko <
> asekretenko@d2iq.com> wrote:
> >>>
> >>> Hi Zoran,
> >>> sure, this particular job can be cancelled/retried.
> >>>
> >>> From looking at the log, there seems to be some infrastructure issue
> on the ASF Jenkins / Git plugin/system/... level:
> >>> the build has yet to reach the build script, and looks stuck at the
> git checkout by the SCM plugin.
> >>> This might be worth investigating on the infra side (especially given
> that Jenkins hadn't aborted the stuck job in 300 minutes; I'm not 100% sure
> if Jenkins is supposed to abort a stuck job at the SCM checkout step,
> though).
> >>>
> >>> On Tue, Oct 27, 2020 at 1:32 PM Zoran Regvart <zo...@regvart.com>
> wrote:
> >>>>
> >>>> Hi Mesos folk,
> >>>> perhaps someone from your side can take a look?
> >>>>
> >>>> thanks!
> >>>>
> >>>> zoran
> >>>> ---------- Forwarded message ---------
> >>>> From: Zoran Regvart <zo...@regvart.com>
> >>>> Date: Tue, Oct 27, 2020 at 11:49 AM
> >>>> Subject: Hung website job
> >>>> To: builds <bu...@apache.org>
> >>>>
> >>>>
> >>>> Hi Builders,
> >>>> seems that a job has hung[1] on the website1 node preventing other
> >>>> website jobs to schedule.
> >>>>
> >>>> Can it be canceled/retried?
> >>>>
> >>>> zoran
> >>>> [1] https://ci-builds.apache.org/job/Mesos/job/Mesos-Websitebot/716/
> >>>> --
> >>>> Zoran Regvart
> >>>>
> >>>>
> >>>> --
> >>>> Zoran Regvart
> >>>
> >>>
> >>>
> >>> --
> >>> --
> >>> Andrei Sekretenko
> >>
> >>
> >>
> >> --
> >> Zoran Regvart
> >
>
>

Re: Hung website job

Posted by Dave Fisher <wa...@apache.org>.
Hi -

I’ve been sledgehammering gitbox with OpenOffice.org migrations. I’m taking a break as I appear to have hit a JBake limit.

Jenkins has a 10 minute timeout git setup.

Regards,
Dave

Sent from my iPhone

> On Nov 2, 2020, at 1:28 PM, Chris Lambertus <cm...@apache.org> wrote:
> 
> Try directing your git checkouts to github. There are some circumstances when gitbox becomes overloaded, let's see if the problem goes away with github.
> 
> 
> 
> 
> 
> 
>> On Nov 2, 2020, at 1:21 PM, Zoran Regvart <zo...@regvart.com> wrote:
>> 
>> Seems that this has happened again with this job:
>> 
>> https://ci-builds.apache.org/job/Lucene/job/Solr-reference-guide-master/404/
>> 
>> as in the previous case there seems to be a snag when checking out from git.
>> 
>> Two cases with the same symptoms, perhaps it's worth investigating?
>> 
>> zoran
>> 
>>> On Tue, Oct 27, 2020 at 2:51 PM Andrei Sekretenko <as...@d2iq.com> wrote:
>>> 
>>> Hi Zoran,
>>> sure, this particular job can be cancelled/retried.
>>> 
>>> From looking at the log, there seems to be some infrastructure issue on the ASF Jenkins / Git plugin/system/... level:
>>> the build has yet to reach the build script, and looks stuck at the git checkout by the SCM plugin.
>>> This might be worth investigating on the infra side (especially given that Jenkins hadn't aborted the stuck job in 300 minutes; I'm not 100% sure if Jenkins is supposed to abort a stuck job at the SCM checkout step, though).
>>> 
>>> On Tue, Oct 27, 2020 at 1:32 PM Zoran Regvart <zo...@regvart.com> wrote:
>>>> 
>>>> Hi Mesos folk,
>>>> perhaps someone from your side can take a look?
>>>> 
>>>> thanks!
>>>> 
>>>> zoran
>>>> ---------- Forwarded message ---------
>>>> From: Zoran Regvart <zo...@regvart.com>
>>>> Date: Tue, Oct 27, 2020 at 11:49 AM
>>>> Subject: Hung website job
>>>> To: builds <bu...@apache.org>
>>>> 
>>>> 
>>>> Hi Builders,
>>>> seems that a job has hung[1] on the website1 node preventing other
>>>> website jobs to schedule.
>>>> 
>>>> Can it be canceled/retried?
>>>> 
>>>> zoran
>>>> [1] https://ci-builds.apache.org/job/Mesos/job/Mesos-Websitebot/716/
>>>> --
>>>> Zoran Regvart
>>>> 
>>>> 
>>>> --
>>>> Zoran Regvart
>>> 
>>> 
>>> 
>>> --
>>> --
>>> Andrei Sekretenko
>> 
>> 
>> 
>> -- 
>> Zoran Regvart
> 


Re: Hung website job

Posted by Chris Lambertus <cm...@apache.org>.
Try directing your git checkouts to github. There are some circumstances when gitbox becomes overloaded, let's see if the problem goes away with github.






> On Nov 2, 2020, at 1:21 PM, Zoran Regvart <zo...@regvart.com> wrote:
> 
> Seems that this has happened again with this job:
> 
> https://ci-builds.apache.org/job/Lucene/job/Solr-reference-guide-master/404/
> 
> as in the previous case there seems to be a snag when checking out from git.
> 
> Two cases with the same symptoms, perhaps it's worth investigating?
> 
> zoran
> 
> On Tue, Oct 27, 2020 at 2:51 PM Andrei Sekretenko <as...@d2iq.com> wrote:
>> 
>> Hi Zoran,
>> sure, this particular job can be cancelled/retried.
>> 
>> From looking at the log, there seems to be some infrastructure issue on the ASF Jenkins / Git plugin/system/... level:
>> the build has yet to reach the build script, and looks stuck at the git checkout by the SCM plugin.
>> This might be worth investigating on the infra side (especially given that Jenkins hadn't aborted the stuck job in 300 minutes; I'm not 100% sure if Jenkins is supposed to abort a stuck job at the SCM checkout step, though).
>> 
>> On Tue, Oct 27, 2020 at 1:32 PM Zoran Regvart <zo...@regvart.com> wrote:
>>> 
>>> Hi Mesos folk,
>>> perhaps someone from your side can take a look?
>>> 
>>> thanks!
>>> 
>>> zoran
>>> ---------- Forwarded message ---------
>>> From: Zoran Regvart <zo...@regvart.com>
>>> Date: Tue, Oct 27, 2020 at 11:49 AM
>>> Subject: Hung website job
>>> To: builds <bu...@apache.org>
>>> 
>>> 
>>> Hi Builders,
>>> seems that a job has hung[1] on the website1 node preventing other
>>> website jobs to schedule.
>>> 
>>> Can it be canceled/retried?
>>> 
>>> zoran
>>> [1] https://ci-builds.apache.org/job/Mesos/job/Mesos-Websitebot/716/
>>> --
>>> Zoran Regvart
>>> 
>>> 
>>> --
>>> Zoran Regvart
>> 
>> 
>> 
>> --
>> --
>> Andrei Sekretenko
> 
> 
> 
> -- 
> Zoran Regvart