Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-11-26
10:38:19 kashyap I even tried to xz-compress the sparse image
10:39:12 kashyap If anyone thinks it's possible ... let me know if you're up for collaborating on it. There are still some slots for 2020 Advent Cal.
10:39:27 kashyap Maybe if you ruthlessly cull man pages, and other stuff out from the filesystem
10:40:32 lyarwood kashyap: you shouldn't need to run ./stack.sh ahead of creating the image tbh, that pulls in loads of stuff that can wait until the guest starts
10:41:02 kashyap lyarwood: Right; I think I tried a few variants
10:42:23 kashyap [And for those who're lurking and wondering ... if you want to contribute an image, here's further instructions: https://www.qemu.org/2020/11/26/qemu-advent-announce/]
10:43:12 kashyap lyarwood: What are the least no. of services we can get away with? [Nova, Glance, Keystone, and Neutron] -- what services can we cull further in there?
10:44:12 kashyap lyarwood: I think it's cheating ... to have an systemd esrvice to just run a simple 'stack.sh' :D But I'll check
11:36:10 openstackgerrit Stephen Finucane proposed openstack/nova stable/train: docs: Rework the PCI passthrough guides https://review.opendev.org/c/openstack/nova/+/764327
11:36:11 openstackgerrit Stephen Finucane proposed openstack/nova stable/train: docs: Change order of PCI configuration steps https://review.opendev.org/c/openstack/nova/+/764328
11:36:12 openstackgerrit Stephen Finucane proposed openstack/nova stable/train: docs: Clarify configuration steps for PF devices https://review.opendev.org/c/openstack/nova/+/764329
11:36:39 stephenfin lyarwood, elod: Would you be able to look at those doc backports real quick? They're clean cherry-picks and doc-only ^
11:37:05 stephenfin I really want the last one, but it's easier bring the other two than resolve merge conflicts IMO
12:14:41 lyarwood tempest.api.compute.admin.test_volume.AttachSCSIVolumeTestJSON.test_attach_scsi_disk_with_config_drive keeps failing at the moment in the ceph job
12:18:11 lyarwood ah right, we try to cleanup the image before the server
12:28:34 openstack Launchpad bug 1905725 in tempest "test_attach_scsi_disk_with_config_drive fails during cleanup when using RBD for Glance and Nova" [Undecided,New]
12:28:34 lyarwood https://review.opendev.org/c/openstack/tempest/+/764337 & https://bugs.launchpad.net/tempest/+bug/1905725 posted
12:32:27 openstackgerrit Lee Yarwood proposed openstack/nova master: DNM test test_attach_scsi_disk_with_config_drive fix https://review.opendev.org/c/openstack/nova/+/764339
12:40:28 gibi fyi, if you miss the CI test result table in the new gerrit UI then http://lists.openstack.org/pipermail/openstack-discuss/2020-November/019051.html
12:42:51 hemanth_n Hi can you take a look at following commits https://review.opendev.org/q/topic:%2522bug/1892361%2522+(status:open) these are backports from ussuri until rocky (patches merged on master and victoria) thanks
12:44:50 lyarwood gibi: noice :)
12:51:19 lyarwood feels like I've gone back in time 10 years, we had a load of greasemonkey scripts when I was in Red Hat support to workaround the awful backend(s) we had to work with at the time
12:52:47 gibi lyarwood: I hacked users script like one script per year. the ability to poke the webpage is useful, but horrible to maintain such script in a changing env.
12:54:09 gibi and for the record I have web development, and I'm bad at it
12:54:20 gibi s/have/hate/
13:14:50 elod stephenfin: sorry, had to run some errands, I've added them to my review queue :)
13:27:40 gibi I got hit by a seemingly new error. grenade fails with to boot an instance with libvirt reporting: Could not open '/dev/sda': Permission denied
13:27:45 gibi https://zuul.opendev.org/t/openstack/build/e74f19d3a4464868b62b5c9dd61911bc/log/controller/logs/screen-n-cpu.txt#7529
13:28:06 gibi logstash shows only a one hit http://logstash.openstack.org/#dashboard/file/logstash.json?query=message%3A%5C%22Could%20not%20open%20'%2Fdev%2Fsda'%3A%20Permission%20denied%5C%22
13:47:00 lyarwood https://review.opendev.org/c/openstack/nova-specs/+/752284 - could use a spec review now that stephenfin is happy if anyone has time this week.
13:47:13 lyarwood gibi: looking
13:47:48 gibi lyarwood: no rush, it seem like a one time thing based on logstash
14:36:39 sean-k-mooney gibi: you script work with codeinjector plugin for firefox too
14:37:02 sean-k-mooney gibi: you need to tick on page load and untick top frame only and it works
14:37:25 sean-k-mooney although it need a refresh
14:38:25 sean-k-mooney nicely done
14:38:42 gibi sean-k-mooney: never used that plugin, tampermonkey works for me in both firefox and chrome :)
14:39:05 sean-k-mooney its the one stephenfin recommened to me a while back
14:39:29 sean-k-mooney i might try tamer monkey but it works so i have not bothered changing
14:39:55 gibi sean-k-mooney: if it works for you in that tool, then I don't think tampermonkey adds too much.
14:39:57 sean-k-mooney gibi: we proaly coudl convert your script into a plugin for gerrit and run it automatically
14:40:23 gibi sean-k-mooney: I don't know how the CI table worked in the old gerrit, I assume that was a plugin that injected the js script from server side
14:40:40 sean-k-mooney gibi: well i have to refewsh as on page load is somethigns before the commensts are there but i could modify it to work more reliably if i really cared
14:40:58 sean-k-mooney gibi: i think it was in java
14:41:08 sean-k-mooney gibi: gerrit used to be written in gwt
14:41:19 sean-k-mooney so i think it was rendered server side
14:41:23 gibi sean-k-mooney: my script should wait for the comments to load with a PerformanceObserver
14:41:39 gibi listening to fetch evnets
14:41:41 gibi events
14:41:51 sean-k-mooney it does seam to work 90% of the time
14:41:52 gibi still there is a bit of sleep added to be sure
14:42:00 sean-k-mooney maybe im just impatiat
14:42:14 sean-k-mooney its working pretty well
14:42:41 sean-k-mooney it does not render the other cis but its good enough for now anyway
14:46:15 sean-k-mooney ah i just need to change if (author == 'Zuul'){
14:46:26 sean-k-mooney to allso look for author that have CI in there name
14:53:15 openstackgerrit Lee Yarwood proposed openstack/nova master: DNM test test_attach_scsi_disk_with_config_drive fix https://review.opendev.org/c/openstack/nova/+/764339
14:57:40 gibi sean-k-mooney: yeah, it is does a lot less than the original CI table
14:57:54 gibi I did a minimum implementation that is already useful for me
14:58:10 sean-k-mooney yep
14:58:48 sean-k-mooney i already refactored it to take the autohr name as an arge just need to get teh list of autor names and loop over all the ci ones
14:58:53 sean-k-mooney althouhg i might no bother
14:59:03 sean-k-mooney at least right now
14:59:25 sean-k-mooney it works fine
14:59:36 elod lyarwood: fyi, I've added a single note to today's nova meeting Stable Branches topic about the live-migration job. feel free to extend the list, either on the wiki or at the meeting :) https://wiki.openstack.org/wiki/Meetings/Nova
15:00:17 sean-k-mooney gibi: its a good basis however for creating plugin to do it server side if we want to long term
15:01:26 gibi sean-k-mooney: I'm not sure how the client side DOM structure relates to the server side java components. I'm happy if my hack helps for the side side work but I have no idea :)
15:01:54 sean-k-mooney gibi: the new gerrrit does not use java anymore
15:02:07 sean-k-mooney its entirely reriten using polomer.js
15:02:20 sean-k-mooney so its all html and javascript calling the rest api
15:03:10 sean-k-mooney its been a wile since i touched polomer but basically it implements web components whcich allows you do define new html tags
15:03:48 sean-k-mooney then you can just place those in the dom and it will invoke the javascript,css and html content of the compnet and redner it
15:05:28 sean-k-mooney i am supper rusty on my web dev but i used to work on https://kato.translatorswb.org/ before i started on openstack
15:05:49 sean-k-mooney it used polemebr.dart + php and twitter bootstracp fo rthe ui
15:06:28 sean-k-mooney ... https://github.com/dart-archive/polymer-dart
15:06:39 sean-k-mooney which was the dart binding to polymer.js
15:08:16 sean-k-mooney anyway it might be a nice side project to work on if i do ill keep you posted
15:09:59 stephenfin Or we finally add the checks plugin to Gerrit. I left a note on the etherpad https://etherpad.opendev.org/p/gerrit-3.2-post-upgrade-notes
15:10:35 sean-k-mooney ya we could but that would require cis to report slightly differently and its still experimental
15:11:01 sean-k-mooney the cheks api is not stable technically in gerrit yet as far as i know
15:22:21 stephenfin it's a plugin so I doubt it ever will be
15:29:09 tacco sean-k-mooney: i foundout something new about my more than 64CPUs issue
15:31:06 tacco i realized with ubuntu i did not had any issues. All 248CPUs up and running with a ubuntu cloud image. strange that fedora and debian buster had the same limitations in here. in debian it 100%is CUSTOM_NR_CPUS_DEFAULT kernel-parameter, could be overrided by grub with nr_cpus=248 but this won't work even if it should work. thats verry strange et all. :D
15:33:16 sean-k-mooney the virt kernels proably have it set smaller or somthign as an optimaisation
15:36:56 tacco yes could be, try to verrify what happens if i build the kernel by myself without that limitations. *crossingFingers*
15:52:22 stephenfin elod: Thank you :)
15:53:48 elod stephenfin: np :)
15:54:39 stephenfin sean-k-mooney: is this a race? If this call fails, we don't roll back the requests on the pools that passed https://github.com/openstack/nova/blob/master/nova/pci/stats.py#L419-L421
15:54:41 elod I didn't see your recheck, so I've double-stamped it :P
15:54:43 stephenfin a bug
15:54:50 stephenfin elod: all good!
15:55:10 elod :]
15:55:10 stephenfin I checked. It was that eventlet issue. Nothing to be worried about, thankfully
15:57:38 gibi nova meeting starts in 2 minutes
16:03:40 sean-k-mooney gibi: for got to add teh flavor metadata thing to the adgenda but we can cover it at the end
16:04:19 sean-k-mooney stephenfin: we operate on a copy in the filter
16:04:32 gibi sean-k-mooney: sure
16:04:44 sean-k-mooney ill take a look after the nova meeting
16:04:44 gibi sean-k-mooney: I will ping you at the end
16:09:37 sean-k-mooney stephenfin: that is only used in the schduler

Earlier   Later