On Fri, Apr 22, 2016 at 01:44:03PM -0600, Kevin Fenzi wrote: > On Fri, 22 Apr 2016 19:41:58 +0200 > Pierre-Yves Chibon <pingou@xxxxxxxxxxxx> wrote: > > > Hi everyone, > > > > Since we are in freeze, I am wondering how people would think about > > upgrading pagure in prod to 2.0. > > > > The list of changes is rather long but there are mostly in the UI, > > the git themselves are un-touched. > > > > So far testing in stg looks fine, I've seen a small glitch in the > > data presented in the UI which is 2 lines of a template to adjust > > (fixed in my git, PR coming). > > > > My idea would be to keep testing things in stg over the week-end and > > roll-out 2.0 (or rather a 2.0.1) on Monday early morning before US > > wakes up to avoid impacting too many people (there might be a few > > minutes of down-time). That of course if people here agree with the > > idea. > > > > So, are you in favor or against this idea? > > Worst case, I'll just wait for after freeze is over :) > > How hard would it be to go back? It would require downgrading the UI and the database (doable since all the migration scripts are using alembic, so we can go up and down) and putting back the forks repos at their former location. Doable, I just hope we don't get to test it :) > Also, I would say avoid the times that branched or rawhide or a RC > compose are running as they pull fedora-pungi-config from pagure. So if they pull via http or ssh, since I suspect they pull from the main repo not a fork, the downtime for this would be about the time it takes to restart apache. Do you know around what time these tasks are run? I seem to remember something like 9am UTC, but I'm not quite sure. > Otherwise +1 I guess. Thanks, I will keep the list updated as to what is happening (upgrade or postpone). Pierre
Attachment:
signature.asc
Description: PGP signature
_______________________________________________ infrastructure mailing list infrastructure@xxxxxxxxxxxxxxxxxxxxxxx http://lists.fedoraproject.org/admin/lists/infrastructure@xxxxxxxxxxxxxxxxxxxxxxx
