I need to upgrade our Red Hat Satellite server in order to get RHEL8 fully working in the lab. I'll need to take an outage to do so. I wanted to get dev input on whether I should pause the smithi queue for the duration or just let RHEL jobs fail. I would hope the outage wouldn't last more than a couple hours. -- David Galloway Systems Administrator, RDU Ceph Engineering IRC: dgalloway
On Wed, Oct 2, 2019 at 6:13 AM David Galloway <dgallowa@redhat.com> wrote:
I need to upgrade our Red Hat Satellite server in order to get RHEL8 fully working in the lab. I'll need to take an outage to do so.
I wanted to get dev input on whether I should pause the smithi queue for the duration or just let RHEL jobs fail. I would hope the outage wouldn't last more than a couple hours.
Rerunning failed RHEL jobs is less disruptive than pausing the queue IMHO. Also vote for doing this on a weekday (not Friday) so it doesn't disrupt over-the-weekend integration tests. -- Patrick Donnelly, Ph.D. He / Him / His Senior Software Engineer Red Hat Sunnyvale, CA GPG: 19F28A586F808C2402351B93C3301A3E258DD79D
On Wed, 2 Oct 2019, David Galloway wrote:
I need to upgrade our Red Hat Satellite server in order to get RHEL8 fully working in the lab. I'll need to take an outage to do so.
I wanted to get dev input on whether I should pause the smithi queue for the duration or just let RHEL jobs fail. I would hope the outage wouldn't last more than a couple hours.
I'd say whatever gets it done most quickly. Right now the queue doesn't have any dev scheduled job except one of mine, which I can reschedule. Pause the queue and get it done? sage
participants (3)
-
David Galloway
-
Patrick Donnelly
-
Sage Weil