Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2024-01-23
16:18:10 bauzas fwiesel: could we be able to see all the job runs in some link ? for example, to see if *all
16:18:20 bauzas changes are getting a FAILURE or not
16:19:01 bauzas formerly, we had a 3rd-party CI link that was telling us that, but this is no longer working :(
16:19:15 bauzas ciwatch.mmedvede.net/project?project=nova
16:19:40 fwiesel bauzas: That's probably doable, but needs to be implemented somehow... The ci itself is not open to the internet and opening that up will be a bit of an effort
16:20:20 bauzas fwiesel: I understand this would be a new /feature/
16:20:22 fwiesel I could definitely put that on the same AWS bucket
16:20:35 bauzas fwiesel: my concern is to not ping you for any failure I'm seeing
16:20:56 bauzas rather if I'm seeing like 90% of the changes getting a failure
16:21:30 fwiesel bauzas: Sure, that can be done. What priority would that have? Compared to commenting on the change request?
16:22:18 bauzas fwiesel: I don't know tbh, this is more about knowing whether we need to ping you or not :p
16:22:26 bauzas but we can start
16:23:03 fwiesel Okay, I'll add it then afterwards.
16:23:11 sean-k-mooney commenting on the change request is the imporant thing
16:23:30 sean-k-mooney having a way to see if the job are generally passing or failing is a nice to have after
16:23:38 bauzas correct
16:23:57 fwiesel I was assuming it, but maybe in the beginning you might trust the ci less and want to see that it produces green builds first
16:24:17 fwiesel Before spamming the changes with messages which are actually faults of the CI
16:24:32 bauzas yeah, most of people would recheck
16:24:36 sean-k-mooney well the ci wont actuly block the patch merging
16:24:46 bauzas but yeah, what sean-k-mooney said
16:24:58 sean-k-mooney so at the start we can just leave it to the reviews to check the ci output when it fails
16:25:15 fwiesel Great, then I'll first do the browsable logs and the comments.
16:25:25 sean-k-mooney +1
16:25:27 bauzas somehow we *eventually* need this for knowing we're done
16:25:40 bauzas but yeah, agreed on the other prios first
16:25:44 bauzas and the reviewable link
16:25:55 bauzas I mean the browsable logs first
16:26:28 bauzas that's still excellent news to know that everything is in place :)
16:26:35 bauzas kudos to you, fwiesel and grandchild
16:26:36 fwiesel "Browseable" means it is okay if it is done in the browser with javascript? Or does it need to be unpacked on the server?
16:26:54 grandchild fwiesel did all the work -- can't take credit :)
16:27:10 bauzas fwiesel: both ? :D
16:27:17 sean-k-mooney normally we convert the logs to html
16:27:30 sean-k-mooney so that we can get clickable links to lines
16:27:38 sean-k-mooney to show this is why x failed
16:27:54 bauzas some others just directly look at the files :p
16:28:09 sean-k-mooney so i would like it to at least be linkable and viewabel
16:28:40 sean-k-mooney the simplest way to do that is normally just enabel directory listing on the server
16:28:40 fwiesel Okay, I'll see what I can do.
16:28:42 bauzas well, on my side, I'd prefer raw files that I could open
16:28:50 bauzas so yeah, unpack first
16:28:57 fwiesel Well, with s3 buckets, there is no directory listing :)
16:28:58 dansmith definitely unpack
16:29:17 dansmith fwiesel: then just generate an HTML file with links to the files and upload that too
16:29:40 fwiesel That I can easily do.
16:29:45 bauzas ++
16:30:14 fwiesel That's from my side.
16:30:28 bauzas thanks a lot
16:30:35 sean-k-mooney cool on a related note where are we with teh hyperv removal
16:30:45 sean-k-mooney gmann: ^ do you have any update on that
16:30:50 bauzas sean-k-mooney: the blueprint was filed and accepted
16:31:03 sean-k-mooney right but we wanted to do it before m2 and we are past that now
16:31:07 bauzas so nothing prevents it to be done, except having someone to do the work
16:31:29 dansmith no reason not to still do it, IMHO
16:31:38 bauzas agreed
16:31:44 bauzas that's what I'm saying
16:31:47 sean-k-mooney right so i was expecting to see patches for this before i went on pto
16:31:49 bauzas everything is paperworked
16:31:54 sean-k-mooney so im try ing to push this forward
16:31:59 bauzas so we just need to merge thinfs
16:32:46 bauzas gmann seems to be off, so I can try to ping him later
16:32:54 sean-k-mooney https://review.opendev.org/c/openstack/nova/+/894466
16:33:03 sean-k-mooney that was the patch but it has nt been updated in a while
16:33:42 bauzas we'll see if gmann can update it shortly
16:33:49 bauzas if not, I can try to take the ball
16:33:56 sean-k-mooney gerrit login is still broken but ya lets at least do a rebase in the next few days
16:34:06 sean-k-mooney cool
16:34:14 bauzas yeah, I just hope I won't get logged out shortly
16:34:52 bauzas ok, last bullet point, then
16:35:04 bauzas #topic Open discussion
16:35:14 bauzas anything anyone ?
16:36:09 bauzas looks not
16:36:16 sean-k-mooney nothing form me
16:36:17 bauzas I save you 23 mins of your time
16:36:22 bauzas thanks all
16:36:24 bauzas #endmeeting
16:36:24 opendevmeet Meeting ended Tue Jan 23 16:36:24 2024 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
16:36:24 opendevmeet Minutes: https://meetings.opendev.org/meetings/nova/2024/nova.2024-01-23-16.01.html
16:36:24 opendevmeet Minutes (text): https://meetings.opendev.org/meetings/nova/2024/nova.2024-01-23-16.01.txt
16:36:24 opendevmeet Log: https://meetings.opendev.org/meetings/nova/2024/nova.2024-01-23-16.01.log.html
16:36:25 sean-k-mooney o/
16:36:28 fwiesel Thanks a lot
16:37:29 sean-k-mooney fwiesel: for refernce we normlaly can view logs in two ways
16:37:37 sean-k-mooney either as a raw file share like https://7abee732bfe2ea059993-afd34e255a9a69e2d84578e615c610d8.ssl.cf1.rackcdn.com/827844/14/check/openstack-tox-pep8/72b9337/
16:37:53 sean-k-mooney or via the zuul application https://zuul.opendev.org/t/openstack/build/72b93377de584b16a0707fc2bcb0158f/logs
16:38:46 fwiesel Thanks, I think, it should be possible to provide both "interfaces"
16:39:45 sean-k-mooney that was from a pep8 job so it does not have devstack logs but ya we just need a way to view the files in the browser without havignto download everything. its also nice ot be able to download the full set but they can get large when often you only need a subset
16:40:47 fwiesel I understand. The idea was not to dump the log as a tarball in general, it was just the fastest way to get it out.
16:43:52 bauzas fwiesel: the shortest for you IMHO is to unpack the tar (or just not tar it) and expose all the files thru an HTML page
16:44:58 fwiesel I think, you'll have that next week.
16:45:13 bauzas (with the right Content-Type in order to open it in my browser instead of downloading it)
16:46:09 bauzas folks, I need to bail out earlier than usual (python meetup in Grenoble)
16:46:13 bauzas see you all tomorrow
16:46:27 fwiesel \o
18:14:17 opendevreview sean mooney proposed openstack/nova master: add initial healthcheck support https://review.opendev.org/c/openstack/nova/+/825015
18:14:18 opendevreview sean mooney proposed openstack/nova master: add healthcheck manager to manager base https://review.opendev.org/c/openstack/nova/+/827844
18:14:18 opendevreview sean mooney proposed openstack/nova master: add healthcheck tracker to nova context https://review.opendev.org/c/openstack/nova/+/829468
18:14:19 opendevreview sean mooney proposed openstack/nova master: add healthcheck utils and constants https://review.opendev.org/c/openstack/nova/+/829469
18:14:19 opendevreview sean mooney proposed openstack/nova master: add healthcheck endpoint to proxy commands https://review.opendev.org/c/openstack/nova/+/830703

Earlier   Later