Monday, 2022-01-17

opendevreviewMerged openstack/tempest master: Add fields in hypervisor schema for 2.33 and 2.53  https://review.opendev.org/c/openstack/tempest/+/73720902:53
*** chandankumar is now known as chkumar|rover06:16
opendevreviewyatin proposed openstack/devstack master: Workaround CentOS 8-stream bug until fixed  https://review.opendev.org/c/openstack/devstack/+/82486206:38
*** jpena|off is now known as jpena08:30
*** akekane_ is now known as abhishekk08:36
opendevreviewyatin proposed openstack/devstack master: Workaround CentOS 8-stream bug until fixed  https://review.opendev.org/c/openstack/devstack/+/82486208:43
opendevreviewBalazs Gibizer proposed openstack/tempest master: Update aggregate tests for bug#1907775  https://review.opendev.org/c/openstack/tempest/+/82164713:47
opendevreviewBalazs Gibizer proposed openstack/tempest master: Introduce @serial test execution decorator  https://review.opendev.org/c/openstack/tempest/+/82173213:47
opendevreviewAlex Yefimov proposed openstack/tempest master: Fix Unittest failures in test_users  https://review.opendev.org/c/openstack/tempest/+/82449113:52
gibigmann: what is the way to bump a dep in tempest?14:41
gibigmann: I did this https://review.opendev.org/c/openstack/tempest/+/821732/8/requirements.txt14:41
gibigmann: but grenade is choking on it https://zuul.opendev.org/t/openstack/build/b7949e580c6048f9a09b798ad2649bdd/log/job-output.txt#2350014:42
gibigmann: global upper constarints are in synch https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L60614:42
artomjparker, so, just realized a thing15:26
artomWe can't have hugepages and file backed memory on the same host15:26
artomhttps://docs.openstack.org/nova/latest/admin/huge-pages.html#enabling-huge-pages-on-the-host15:26
artomMeaning if we're going to run the hugepages tests upstream (which I think we should), we'll need to either split them off into their own job, or split the file backed memory tests into their own job15:27
artomI suppose the former is probably better?15:30
gmanngibi: that is right way, checking the error15:57
gmanngibi: got the problem. it try to install tempest at system level and greande job using stable/xena devstack and constraints. https://zuul.opendev.org/t/openstack/build/b7949e580c6048f9a09b798ad2649bdd/log/job-output.txt#2343616:16
gmanngibi: we should not install tempest at system level in stable. I will fix the configuration for stable which we forgot while cutting stable branch(I will add this step in QA release wiki page to take care of it in future)16:17
gibigmann: awesome, thank you!16:24
opendevreviewGhanshyam proposed openstack/devstack stable/xena: Stop installing Tempest at system wide for stable branch  https://review.opendev.org/c/openstack/devstack/+/82499116:26
opendevreviewGhanshyam proposed openstack/devstack stable/wallaby: Stop installing Tempest at system wide for stable branch  https://review.opendev.org/c/openstack/devstack/+/82495616:26
gmanngibi: ^^, after this requirement things will work fine16:27
gmannkopecmartin: frickler  quick one to review  https://review.opendev.org/q/I65304f66b3fe9c372fad8d4b521d0219833c1bf116:27
gibigmann: ack16:27
*** jpena is now known as jpena|off17:29
jparkerartom_: yea I agree making a dedicated hugepage job makes sense17:58
jparkerartom_: I put up the tests for the review, but I want to get another commit in place to setup the environment.17:59
jparkerartom_: I guess I could split the jobs in that commit as well?17:59
artom_jparker, yeah, that would make sense18:07
opendevreviewMerged openstack/devstack master: Workaround CentOS 8-stream bug until fixed  https://review.opendev.org/c/openstack/devstack/+/82486218:17
jparkerartom_: does the zuul job(s) in whitebox inherent from another project that sets file backed memory?19:23
artom_jparker, no, the tests do that on the fly19:33
artom_But if we deploy with hugepages and the run them, I'm assuming nova-compute won't start, or something else will break19:33
jparkerartom_: oh now I see it ok19:35
jparkerartom_: and how much memory is allocated to the compute's upstream?19:36
artom_jparker, err, lemme find that - IIRC, it's available in the project-config repo's YAML somewhere19:43
artom_Actually, an easier way might be to look at the ansible inventory of a recent job19:44
artom_Because the RAM will depend on the exact flavor that our nodepool label uses19:45
* jparker checking now19:45
artom_jparker, https://zuul.opendev.org/t/openstack/build/abdea27bf6d44d81856998aca2126f79/log/zuul-info/host-info.compute.yaml#39519:46
artom_32 GB, apparently19:46
jparkerartom_: ok so maybe 4GB of 2MB and 4GB of 1MB?19:47
jparkeror maybe 2GB for each size19:48
artom__jparker, I suppose, though the less the better, as those pages are only usable by processes that explicitly request them19:49
artom__So they're unavailable for the rest of the system19:50
jparkerartom__: ack ok I'll start smaller first19:50
artom__So only the minimum of what's needed to run a small guest backed by huge pages19:50
*** artom__ is now known as artom19:50
opendevreviewJames Parker proposed openstack/whitebox-tempest-plugin master: [WIP] Add multiple hugepage size job  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/82501120:13
opendevreviewJames Parker proposed openstack/whitebox-tempest-plugin master: [WIP] Add multiple hugepage size job  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/82501120:26
*** kopecmartin8 is now known as kopecmartin20:47
opendevreviewGhanshyam proposed openstack/tempest master: Revert "Make centos8 stream job non voting for bug#1957941"  https://review.opendev.org/c/openstack/tempest/+/82496222:09
opendevreviewJames Parker proposed openstack/whitebox-tempest-plugin master: [WIP] Add multiple hugepage size job  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/82501122:43

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!