On Jul 16 11:37, Praveen Kumar wrote:
> On Mon, Jul 16, 2018, 11:33 AM Lalatendu Mohanty <lmohanty@redhat.com>
> wrote:
>
> > Hi,
> >
> > On Mon, Jul 16, 2018 at 5:31 AM, <ci@centos.org> wrote:
> >
> >> The build has failed. Check the build URL for failure logs:
> >> https://ci.centos.org/job/minishift-nightly-centos/218/
> >
> >
> > Our nightly builds are failing because of a CICO issue.  Here is the error
> > from last test run.
> >
> > *00:01:21* ++ cico node get -f value -c ip_address -c comment --retry-count 12*00:01:25* Starting new HTTP connection (1): admin.ci.centos.org*00:01:28* Resetting dropped connection: admin.ci.centos.org*00:01:30* + [[ '' != '' ]]*00:01:30* + echo 'CICO: Unable to get Node.'*00:01:30* CICO: Unable to get Node.*00:01:30* + exit 1*00:01:30* Build step 'Execute shell' marked build as failure*00:01:30* Email was triggered for: Failure - Any*00:01:30* Sending email for trigger: Failure - Any
> > xxxxxx
> > *00:01:31* Finished: FAILURE
> >
> >
> > We have been seeing issues like this for our nightly runs. Refer [1]. How
> > can we fix the issue from CentOS ci side?
> >
>
> Most of time this is cause of failure, looks like a certain time nodes are
> not allocated, better to ask Fabian or any other person who is responsible
> for ci infra.
>
>
> > [1] https://ci.centos.org/job/minishift-nightly-centos/218/
> >
> > -Lala
> >
> >
> > _______________________________________________
> > Ci-users mailing list
> > Ci-users@centos.org
> > https://lists.centos.org/mailman/listinfo/ci-users
> >

You have a few nightly jobs that all start at midnight. Recall that
duffy has a rate limit of 6 sessions over 5 minutes.

If you can stagger some of these nightly jobs so that they don't run at
once you should be able to get machines.

The appropriate cron line looks like this, and will run the job at a
random time during the midnight hour:

H 0 * * *

--Brian

_______________________________________________
Ci-users mailing list
Ci-users@centos.org
https://lists.centos.org/mailman/listinfo/ci-users