[openstack-dev] [tripleo][puppet] Hold off on approving patches until further notice
pabelanger at redhat.com
Fri Aug 11 12:59:07 UTC 2017
On Thu, Aug 10, 2017 at 10:32:25PM -0600, Alex Schultz wrote:
> On Thu, Aug 10, 2017 at 11:12 AM, Paul Belanger <pabelanger at redhat.com> wrote:
> > On Thu, Aug 10, 2017 at 07:03:32AM -0600, Alex Schultz wrote:
> >> FYI,
> >> The gates are hosed for a variety of reasons and we can't get
> >> critical patches merged. Please hold off on rechecking or approving
> >> anything new until further notice. We're hoping to get some of the
> >> fixes for this merged today. I will send a note when it's OK to merge
> >> again.
> >>  https://bugs.launchpad.net/tripleo/+bug/1709428
> >>  https://bugs.launchpad.net/tripleo/+bug/1709327
> > So far, these are the 3 patches we need to land today:
> > Exclude networking-bagpipe from dlrn
> > - https://review.openstack.org/491878
> > Disable existing repositories in tripleo-ci
> > - https://review.openstack.org/492289
> > Stop trying to build networking-bagpipe with DLRN
> > - https://review.openstack.org/492339
> > These 3 fixes will take care of the large amount of gate resets tripleo is
> > currently seeing. Like Alex says, please try not to approve / recheck anything
> > until we land these.
> Ok so we've managed to land patches to improve the reliability.
> https://review.openstack.org/492339 - merged
> https://review.openstack.org/491878 - still pending but we managed to
> get the package fixed so this one is not as critical anymore
> https://review.openstack.org/491522 - merged
> https://review.openstack.org/492289 - merged
> We found that the undercloud-container's job is still trying to pull
> from buildlogs.centos.org, and I've proposed a fix
> I've restored (and approved) previously approved patches that have a
> high/critical bug or a FFE approved blueprint associated.
> It should be noted that the following patches for tripleo do not have
> a bug or bp reference so they should be updated prior to being
> For tripleo patches, please refer to Emilien's email about the RC
> schedule with includes these rules about what patches should be
> merged. Please be careful on rechecks and check failures. Do not
> blindly recheck. We have noticed some issues with citycloud nodes, so
> if you spot problems with specific clouds please let us know so we can
> track these and work with infra on it.
>  http://lists.openstack.org/pipermail/openstack-dev/2017-August/120806.html
For today I'm going to keep an eye on elastic-recheck and see what is still
failing in the gate. I agree, for the most part we seem to be looking good on
infrastructure issues but I think the container jobs are still failing too much
for my liking.
Lets see what happens today.
More information about the OpenStack-dev