Open Stack

I agree we need to separate out the different CI jobs into their own accounts.   We'll start on that and I’ll get back to you

Thanks,
Rich

From: Edgar Magana <edgar.magana at workday.com<mailto:edgar.magana at workday.com>>
Reply-To: "Announcements for third party CI operators." <third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>>
Date: Thursday, November 20, 2014 at 4:45 PM
To: "third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>" <third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>>
Subject: Re: [Third-party-announce] Cisco CSR1KV CI Failing and recheck commands

Rich,

My strong suggestion is to separate Cisco CI into multiple accounts. You can request new accounts for these different jobs/technologies:
http://ci.openstack.org/third_party.html#requesting-a-service-account

It will provide a more granular visibility for everybody as well as a easier management when one of those jobs got broken again (hopefully not very often).

On the recheck commands, please update the wiki page with the corresponding strings:
https://wiki.openstack.org/wiki/Neutron_Plugins_and_Drivers#Existing_Plugin_and_Drivers

Unfortunately, I feel uncomfortable to re-enable a CI system that is incomplete at this moment. I believe that the above suggestion will help in many ways avoiding this situation again.

Thanks,

Edgar

From: "Richard Winters (riwinter)" <riwinter at cisco.com<mailto:riwinter at cisco.com>>
Reply-To: "third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>" <third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>>
Date: Thursday, November 20, 2014 at 1:04 PM
To: "third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>" <third-party-announce at lists.openstack.org<mailto:third-party-announce at lists.openstack.org>>
Subject: Re: [Third-party-announce] Cisco CSR1KV CI Failing and recheck commands

Hello Anita, Edgar

I’m working with Nikolay and Dane on Cisco’s CI stability problem.

We are currently evaluating the failures on the CSR1KV CI and implementing corrective actions to stabilize it.  While this activity is progressing we have internally disabled both the CSR1KV as well as the VPNaaS jobs from running.

At this time you could reenable the Cisco CI to pick up the remaining jobs (apic-nexus, ml2-multinode-nexus, ml2-nexus and n1kv_neutron) which we feel have been running reliably.  Once we have identified and corrected all the problems with the CSR1KV/VPNaaS jobs we’ll go through any necessary procedures to reintroduce them.

Also, AFAIK the Cisco recheck command is available and hasn’t changed since it was first implemented.   Let me know if your looking for a change in the way it currently works.

Thanks
Rich

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/third-party-announce/attachments/20141120/31aed477/attachment.html>