[Ci-users] [ci-users] What is the idea --retry-count ?

David Moreau Simard dms at redhat.com
Wed Apr 19 14:39:10 UTC 2017


The API will return the same "error" if there is no inventory left
available or if you have reached rate limiting.

It's typically rare to be out of inventory (unless there's some
issue), so you're probably hitting the rate limit.
There's different rate limiting in place to prevent a tenant to
exhaust the entire pool of nodes -- I think, for example, there is an
"amount of nodes" per 5 minutes window or something like that.

Try to get in touch with the ci.centos.org admins to see if they can
help you figure out what's happening.

David Moreau Simard
Senior Software Engineer | Openstack RDO

dmsimard = [irc, github, twitter]


On Wed, Apr 19, 2017 at 12:18 AM, Budh Gurung <bgurung at redhat.com> wrote:
> Hi,
>
> With the first incidence of "node not available / No inventory available"
> around 2 weeks ago (3/4th April), we updated our Job configurations [1]
> to have a retry-count of 6 which sums up with default wait of 10 seconds
> to 1 minute.
>
> We thought that would be enough.
>
> But now, again we are intermittently experiencing similar incidence [2] even
> having a retry-count to 6.
>
> I am curious here, what is the ideal --retry-count here.
>
>
> [1]
> https://github.com/minishift/minishift-ci-jobs/blob/master/minishift-ci-index.yaml#L66
> [2] https://ci.centos.org/job/minishift/94/console
>
>
> Regards,
> Budh Ram Gurung
> Software Engineer - Devtools
>
> _______________________________________________
> Ci-users mailing list
> Ci-users at centos.org
> https://lists.centos.org/mailman/listinfo/ci-users
>



More information about the Ci-users mailing list