| *** Lucas_Gray has joined #openstack-ironic | 00:05 | |
| *** Lucas_Gray has quit IRC | 00:12 | |
| *** Lucas_Gray has joined #openstack-ironic | 00:14 | |
| *** ociuhandu has joined #openstack-ironic | 01:01 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Follow up of fix uefi jobs with ovmf native ubuntu package https://review.opendev.org/739244 | 01:03 |
|---|---|---|
| *** ociuhandu has quit IRC | 01:05 | |
| *** cdearborn has quit IRC | 01:15 | |
| *** diurnalist has quit IRC | 01:56 | |
| *** cp- has quit IRC | 01:57 | |
| *** cp- has joined #openstack-ironic | 01:59 | |
| *** diurnalist has joined #openstack-ironic | 02:06 | |
| *** diurnalist has quit IRC | 02:11 | |
| *** rcernin has quit IRC | 02:22 | |
| *** rcernin has joined #openstack-ironic | 02:22 | |
| *** rloo has quit IRC | 02:29 | |
| *** diurnalist has joined #openstack-ironic | 03:00 | |
| *** diurnalist has quit IRC | 03:05 | |
| *** dsneddon has quit IRC | 03:06 | |
| *** yaawang has quit IRC | 03:17 | |
| *** yaawang has joined #openstack-ironic | 03:21 | |
| *** diurnalist has joined #openstack-ironic | 03:55 | |
| *** diurnalist has quit IRC | 04:00 | |
| *** Lucas_Gray has quit IRC | 04:13 | |
| *** diurnalist has joined #openstack-ironic | 04:47 | |
| *** diurnalist has quit IRC | 04:52 | |
| *** abdysn has joined #openstack-ironic | 05:18 | |
| *** diurnalist has joined #openstack-ironic | 05:24 | |
| *** yaawang has quit IRC | 05:27 | |
| *** yaawang has joined #openstack-ironic | 05:27 | |
| *** diurnalist has quit IRC | 05:29 | |
| *** janders has joined #openstack-ironic | 05:30 | |
| arne_wiebalck | Good morning, ironic! | 06:01 |
| janders | g'day Arne | 06:02 |
| *** brtknr has quit IRC | 06:15 | |
| *** diurnalist has joined #openstack-ironic | 06:21 | |
| *** brtknr has joined #openstack-ironic | 06:24 | |
| prometheanfire | dtantsur|afk: you every make a review for jsonpatch===1.25 causes test failures for ironic, fyi https://zuul.opendev.org/t/openstack/build/ff068dc014be42aabde51f95c8c9828e | 06:24 |
| prometheanfire | dtantsur|afk: asking for https://review.opendev.org/739652 | 06:24 |
| patchbot | patch 739652 - requirements - manual generate-constraints run - 1 patch set | 06:24 |
| *** diurnalist has quit IRC | 06:26 | |
| rpittau | prometheanfire: I think that issue is not related to jsonpatch, was fixed by https://review.opendev.org/739418 | 07:06 |
| patchbot | patch 739418 - ironic - Use unittest mock instead of 3rd party mock in tes... (MERGED) - 1 patch set | 07:06 |
| rpittau | good morning ironic! o/ | 07:06 |
| *** dtantsur|afk is now known as dtantsur | 07:18 | |
| dtantsur | morning ironic | 07:18 |
| dtantsur | prometheanfire: what rpittau said | 07:18 |
| openstackgerrit | zhufl proposed openstack/ironic master: Fix missing print format in log message https://review.opendev.org/739667 | 07:19 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent master: Use ubuntu focal for ipa src https://review.opendev.org/725794 | 07:21 |
| dtantsur | huh, what's wrong with the ussuri CI again? | 07:21 |
| rpittau | heh | 07:22 |
| iurygregory | good morning arne_wiebalck rpittau dtantsur and Ironic o/ | 07:22 |
| dtantsur | \o | 07:22 |
| rpittau | hey iurygregory :) | 07:22 |
| dtantsur | we've merged quite a few useful things to ironic train, does anyone want to propose a release? | 07:26 |
| dtantsur | same with IPA once https://review.opendev.org/739553 merges | 07:26 |
| patchbot | patch 739553 - ironic-python-agent (stable/train) - Fix serializing ironic-lib exceptions - 1 patch set | 07:26 |
| rpittau | dtantsur: yep | 07:26 |
| dtantsur | let's do it as a warm-up :) then we can look into master releases | 07:28 |
| rpittau | sounds good :) | 07:29 |
| rpittau | also good for what I'm writing on how to propose a release | 07:29 |
| dtantsur | yup | 07:29 |
| rpittau | so I think release notes and patches are ok, I'll go directly to releases | 07:30 |
| arne_wiebalck | hey iurygregory rpittau dtantsur o/ | 07:31 |
| rpittau | hey arne_wiebalck :) | 07:31 |
| *** diurnalist has joined #openstack-ironic | 07:32 | |
| rpittau | wow there's a bunch of stuff for train | 07:33 |
| *** diurnalist has quit IRC | 07:37 | |
| *** dsneddon has joined #openstack-ironic | 07:39 | |
| janders | hi team! :) | 07:40 |
| janders | do we have plans to support Bifrost on FC32? | 07:40 |
| rpittau | janders: I think we do :) | 07:40 |
| janders | in release notes I see FC30 is supported.. trying to get it up on my FC32 and hitting minor issues (renamed packages) which I am happy to fix | 07:41 |
| janders | ok! great | 07:41 |
| rpittau | janders: the point now is that the base os in CI is Fedora 30 | 07:41 |
| rpittau | janders: patches are welcome! :) | 07:41 |
| janders | :) | 07:42 |
| *** dsneddon has quit IRC | 07:44 | |
| dtantsur | janders: hey! we do intend that, but nobody has tried yet. patches welcome (or just use CentOS 8 which is supported) | 07:49 |
| iurygregory | F32 you need to remove one package from the list of fedora packages | 07:49 |
| iurygregory | I've tired once | 07:49 |
| rpittau | I also tried a couple times, didn't propose any change as CI is on FC30 still | 07:50 |
| iurygregory | yup | 07:50 |
| janders | libselinux-python is the problem package, right? | 07:55 |
| *** rcernin has quit IRC | 07:57 | |
| iurygregory | janders, yup =) | 07:57 |
| janders | iurygregory thank you | 07:57 |
| iurygregory | you can just remove and it should just work =) | 07:57 |
| iurygregory | remove the line hehe XD | 07:58 |
| janders | does equivalent package get pulled in automatically? | 07:58 |
| iurygregory | python3-libselinux should take care | 07:58 |
| iurygregory | I think we have on the list | 07:59 |
| janders | I will look through my mods when the playbook concludes - I think there are two files, one is exactly as you described, another needed a swap | 07:59 |
| iurygregory | I checked with a fedora package maintainer | 07:59 |
| iurygregory | I think I only had to change in the bifrost-ironic-install /defaults/required_defaults_Fedora.yml | 08:01 |
| janders | here are my two mods on the working copy | 08:01 |
| janders | diff --git a/playbooks/roles/bifrost-create-vm-nodes/defaults/required_defaults_Fedora.yml b/playbooks/roles/bifrost-create-vm-nodes/defaults/required_defaults_Fedora.ymlindex 6a5ef12..719525e 100644--- a/playbooks/roles/bifrost-create-vm-nodes/defaults/required_defaults_Fedora.yml+++ | 08:02 |
| janders | b/playbooks/roles/bifrost-create-vm-nodes/defaults/required_defaults_Fedora.yml@@ -11,4 +11,5 @@ required_packages: - libvirt-daemon - pkgconfig - libvirt-devel- - libselinux-python+# - libselinux-python+ - python3-libselinuxdiff --git a/playbooks/roles/bifrost-ironic-install/defaults/required_defaults_Fedora.yml | 08:02 |
| janders | b/playbooks/roles/bifrost-ironic-install/defaults/required_defaults_Fedora.ymlindex 21a7be4..9a83eb6 100644--- a/playbooks/roles/bifrost-ironic-install/defaults/required_defaults_Fedora.yml+++ b/playbooks/roles/bifrost-ironic-install/defaults/required_defaults_Fedora.yml@@ -14,7 +14,7 @@ required_packages: - ipxe-bootimgs - kpartx - | 08:02 |
| janders | libffi-devel- - libselinux-python+# - libselinux-python - libxml2-devel - libxslt-devel - mariadb-server[and512@xps bifrost]$ | 08:02 |
| janders | oops thats not very readable | 08:02 |
| iurygregory | hehehe | 08:02 |
| iurygregory | paste.openstack.org | 08:02 |
| * iurygregory facepalm just saw that the meetup was yesterday =( | 08:03 | |
| janders | http://paste.openstack.org/show/795596/ | 08:07 |
| janders | it seems there are two yaml files - one in bifrost-create-vm-nodes and one in bifrost-ironic-install | 08:07 |
| iurygregory | yeah there are two, I only had to change the /bifrost-ironic-install/defaults/required_defaults_Fedora.yml | 08:08 |
| janders | right - and with the latter, there is no need to add the new package, just remove the old | 08:08 |
| janders | with the former, I didn't see anything FC32 like so I added python3-libselinux. Unsure what it would do without, I might test that | 08:09 |
| iurygregory | you just need to remove/comment the line, the thing is that the package doesn't exist in F32 so it fails during bifrost-ironic-install | 08:10 |
| janders | ack! thank you iurygregory | 08:10 |
| iurygregory | and python3-libselinux is required already https://github.com/openstack/bifrost/blob/4e83214d37adc213138d240c55e063508c26a19e/playbooks/roles/bifrost-ironic-install/defaults/required_defaults_Fedora.yml#L28 | 08:11 |
| dtantsur | will the change work for 30 as well? would be handy not to duplicate the packages | 08:14 |
| iurygregory | dtantsur, I haven't tested with F30... | 08:14 |
| *** lucasagomes has joined #openstack-ironic | 08:14 | |
| janders | good question dtantsur... I was wondering if we need a conditional | 08:14 |
| * dtantsur tries in podman | 08:14 | |
| dtantsur | a handy trick for testing various OS, e.g. F30: podman run -it fedora:30 /bin/bash | 08:15 |
| iurygregory | if python3-libselinux pulls libselinux-python in F30 we should be fine | 08:15 |
| *** rcernin has joined #openstack-ironic | 08:15 | |
| iurygregory | let me check in my F30 NUC | 08:15 |
| dtantsur | python3-libselinux seems to be a thing in F30 | 08:16 |
| dtantsur | libselinux-python3 is already an alias | 08:16 |
| rpittau | I geuss firing it up in the CI could give us a good answer :) | 08:16 |
| dtantsur | janders: you can just replace libselinux-python with python3-libselinux | 08:17 |
| dtantsur | actually, libselinux-python is for Python 2, we don't need it :) | 08:17 |
| janders | well spotted! :) | 08:18 |
| iurygregory | burn with fire | 08:18 |
| janders | I was considering looking up the commit that added that line to get context, but in this case I won't | 08:18 |
| janders | kill -10! :) | 08:18 |
| dtantsur | I suspect it was cargo-culted from python 2 times | 08:19 |
| iurygregory | yup | 08:19 |
| iurygregory | and since F30 has py27 support they have the lib | 08:19 |
| dtantsur | right | 08:20 |
| iurygregory | is just me or is strange that the huawei CI provides results very fast? | 08:23 |
| dtantsur | it surprises me too | 08:24 |
| *** rcernin has quit IRC | 08:29 | |
| *** Lucas_Gray has joined #openstack-ironic | 08:29 | |
| *** derekh has joined #openstack-ironic | 08:30 | |
| *** antonym has quit IRC | 08:38 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Add documentation on how to propose a release https://review.opendev.org/739678 | 08:38 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Add documentation on how to propose a release https://review.opendev.org/739678 | 08:40 |
| *** bfournie has quit IRC | 08:41 | |
| *** rcernin has joined #openstack-ironic | 08:44 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent master: Refactor part of image module https://review.opendev.org/739251 | 08:44 |
| *** rcernin has quit IRC | 08:45 | |
| *** diurnalist has joined #openstack-ironic | 08:46 | |
| *** janders has quit IRC | 08:48 | |
| *** priteau has joined #openstack-ironic | 08:49 | |
| *** diurnalist has quit IRC | 08:51 | |
| *** janders has joined #openstack-ironic | 08:52 | |
| dtantsur | rpittau: to follow up on https://review.opendev.org/#/c/739439/ do you think a reno is necessary? we don't usually tend to have them for cosmetic changes. | 08:54 |
| patchbot | patch 739439 - ironic - Do not put the whole JSON error from the agent in ... - 1 patch set | 08:54 |
| rpittau | dtantsur: I was thinking if we want to backport that maybe we need a reno even if it's mostly a cosmetic change, since there is the fix for the error detection for TypeError | 08:56 |
| *** yaawang has quit IRC | 08:57 | |
| rpittau | dtantsur: I trust you and TheJulia though, if you think it's not necessary I'm ok with that, maybe I'm just being too strict :) | 08:57 |
| *** yaawang has joined #openstack-ironic | 08:57 | |
| dtantsur | I don't care much, just don't want to respin the patch given the CI state :) | 08:57 |
| openstackgerrit | Merged openstack/ironic-python-agent stable/ussuri: Fix serializing ironic-lib exceptions https://review.opendev.org/739552 | 08:58 |
| rpittau | dtantsur: I get that, I'll approve it, if it's needed we can add one as a follow up | 08:58 |
| openstackgerrit | Merged openstack/ironic-python-agent stable/train: Fix serializing ironic-lib exceptions https://review.opendev.org/739553 | 08:58 |
| * dtantsur has just learned oslo.config has an 'advanced' flag for options | 08:59 | |
| *** k_mouza has joined #openstack-ironic | 08:59 | |
| rpittau | uuuuh we can ask a release for ipa train ? | 08:59 |
| dtantsur | yup | 09:02 |
| *** bfournie has joined #openstack-ironic | 09:02 | |
| *** diurnalist has joined #openstack-ironic | 09:02 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Add documentation on how to propose a release https://review.opendev.org/739678 | 09:04 |
| *** diurnalist has quit IRC | 09:07 | |
| dtantsur | TheJulia: it doesn't seem like oslo.config supports groups with freeform options :( I've updated https://storyboard.openstack.org/#!/story/1618014 to use MultiOpt which looks a bit nicer | 09:08 |
| *** hjensas has quit IRC | 09:13 | |
| dtantsur | I don't want to bring bad news, but https://zuul.openstack.org/builds?job_name=ironic-tempest-ipa-partition-uefi-pxe_ipmitool | 09:15 |
| rpittau | I was looking at that too... | 09:15 |
| iurygregory | ouch | 09:16 |
| dtantsur | what I've seen was kernel not booting, you? | 09:16 |
| rpittau | same | 09:17 |
| rpittau | it fails on second pass | 09:17 |
| rpittau | and actually I see kernel panic too | 09:17 |
| rpittau | [ 25.864622] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) | 09:18 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 09:19 |
| rpittau | it's interesting as it's not always | 09:19 |
| * rpittau meditates on writing a guide on the meaning of 'interesting' | 09:20 | |
| *** yaawang has quit IRC | 09:23 | |
| dtantsur | we use "interesting" in the ironic sense! | 09:25 |
| *** yaawang has joined #openstack-ironic | 09:25 | |
| *** ricolin has quit IRC | 09:27 | |
| *** hjensas has joined #openstack-ironic | 09:34 | |
| *** diurnalist has joined #openstack-ironic | 09:38 | |
| *** ociuhandu has joined #openstack-ironic | 09:42 | |
| *** diurnalist has quit IRC | 09:43 | |
| rpittau | I was starting to be worried it was the change for the ovmf issue, but it's not, it's failing also in ussuri and we still have the xenial trick in palce | 09:43 |
| rpittau | place* | 09:43 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Do not use 'sudo pip install' when venv is used https://review.opendev.org/739693 | 09:44 |
| dtantsur | rpittau: FYI ^^ | 09:44 |
| rpittau | ah lol | 09:44 |
| rpittau | we should probably use the module command | 09:45 |
| dtantsur | this code will be much nicer when we drop support for non-venv installations | 09:46 |
| rpittau | yeah | 09:46 |
| dtantsur | by the way, we can probably do it after the upcoming release | 09:47 |
| dtantsur | since the deprecation landed in Ussuri | 09:47 |
| rpittau | oh yeah, we can release bifrost too I guess, do we want to wait for this to merge ? | 09:47 |
| dtantsur | we should release bifrost actually. I'd like to get this in if possible. | 09:48 |
| rpittau | ok, let's give it some time | 09:48 |
| dtantsur | the bifrost CI is fast if not broken :) | 09:49 |
| rpittau | dtantsur: did we change ipa ramdisk recently? | 09:57 |
| rpittau | oh but even that, it doesn't make sense, the other uefi job is working fine | 09:57 |
| dtantsur | rpittau: we do change it from time to time | 09:57 |
| rpittau | ok I think I know what's happening | 09:59 |
| *** Qianbiao has joined #openstack-ironic | 10:02 | |
| *** k_mouza has quit IRC | 10:02 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Follow up of fix uefi jobs with ovmf native ubuntu package https://review.opendev.org/739244 | 10:05 |
| rpittau | dtantsur ^ this now includes also the memory increase for the other uefi job that was not included in the original patch | 10:05 |
| rpittau | that should fix the issue | 10:05 |
| rpittau | the dib based image requires much more space | 10:06 |
| rpittau | we should probably apply the same to ussuri | 10:06 |
| rpittau | let's see how the CI goes | 10:06 |
| dtantsur | yup | 10:08 |
| *** k_mouza has joined #openstack-ironic | 10:10 | |
| iurygregory | so if I got right some dib jobs work fine with 3GB but other jobs we need 4GB? | 10:13 |
| dtantsur | UEFI jobs, specifically | 10:13 |
| *** diurnalist has joined #openstack-ironic | 10:14 | |
| iurygregory | gotcha | 10:14 |
| iurygregory | =) | 10:14 |
| *** hjensas is now known as hjensas|afk | 10:15 | |
| *** diurnalist has quit IRC | 10:19 | |
| janders | sometimes it seems to me that the "U" in UEFI stands for Unusable... | 10:21 |
| *** yaawang has quit IRC | 10:21 | |
| janders | but - I hit a particularly hard to run batch of UEFI kit on a past project, so I'm totally biased :) | 10:21 |
| *** yaawang has joined #openstack-ironic | 10:23 | |
| dtantsur | heh | 10:26 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Do not validate driver on changing non-driver fields https://review.opendev.org/739699 | 10:26 |
| janders | I'm trying to progress my FC32 bifrost install, hitting issues with DB stuff: | 10:26 |
| janders | http://paste.openstack.org/show/795599/ | 10:26 |
| janders | real weird thing is it seems to me that db-sync is doing it's job, DB and tables do exist | 10:27 |
| janders | and schema creation fails with no error | 10:27 |
| janders | have you seen anything similar? | 10:27 |
| janders | I'd really appreciate a pointer or two :) | 10:28 |
| janders | In this particular output I set ignore_errors to true for DB Schema tasks, just to see what happens. Prior to that it would fail out a task or two earlier than in this output. | 10:29 |
| dtantsur | I wonder if we need more hacks for F32 | 10:32 |
| janders | yeah I'd love to spin up a C8 somewhere and see if I hit any of these | 10:33 |
| janders | (or a clean FC32) | 10:33 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Do not put the whole JSON error from the agent in last_error https://review.opendev.org/739439 | 10:37 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allow deleting nodes with a broken driver https://review.opendev.org/739702 | 10:38 |
| dtantsur | janders: you need to check the mysql socket path on your system | 10:41 |
| dtantsur | we currently assume /var/lib/mysql/mysql.sock for RH systems | 10:41 |
| janders | [root@xps ~]# ls -alZ /var/lib/mysql/mysql.sock srwxrwxrwx. 1 mysql mysql system_u:object_r:mysqld_var_run_t:s0 0 Jul 7 20:23 /var/lib/mysql/mysql.sock[root@xps ~]# | 10:42 |
| janders | that does exist on mine, too | 10:42 |
| janders | \/etc/my.cnf.d/mariadb-server.cnf:socket=/var/lib/mysql/mysql.sock | 10:43 |
| janders | interestingly it's intermittent | 10:44 |
| janders | now it passed the task with DB connection | 10:44 |
| janders | and failed on inspector DB creation | 10:45 |
| dtantsur | yup, noticed that as well | 10:45 |
| janders | http://paste.openstack.org/show/795603/ | 10:46 |
| janders | at least we have an error this time | 10:46 |
| iurygregory | janders, I will give a try on a F32 server after lunch to see | 10:52 |
| janders | great, thank you iurygregory | 10:52 |
| iurygregory | yw | 10:52 |
| janders | I will also try to test on a clean machine when I get more hardware (hopefully later this week) | 10:53 |
| janders | meanwhile added another ignore_errors and will kick off the final attempt for the day I suppose | 10:53 |
| janders | it really feels like something related to hardware somehow. The intermittent weirdness.. The laptop's running fine other than that though! | 10:54 |
| dtantsur | clean machine ++ | 10:54 |
| *** priteau has quit IRC | 11:00 | |
| openstackgerrit | Merged openstack/bifrost master: Do not use 'sudo pip install' when venv is used https://review.opendev.org/739693 | 11:02 |
| *** tkajinam has quit IRC | 11:04 | |
| *** ociuhandu has quit IRC | 11:05 | |
| *** ociuhandu has joined #openstack-ironic | 11:06 | |
| arne_wiebalck | TheJulia: I had a look at the memory guard patch and left a question on the review. | 11:18 |
| openstackgerrit | Derek Higgins proposed openstack/ironic master: CI: Make ipv6 job to voting https://review.opendev.org/739715 | 11:25 |
| *** hjensas|afk is now known as hjensas | 11:30 | |
| openstackgerrit | Merged openstack/ironic-python-agent stable/ussuri: Limit Inspection->Lookup->Heartbeat lag https://review.opendev.org/739569 | 11:49 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost stable/ussuri: Do not use 'sudo pip install' when venv is used https://review.opendev.org/739718 | 11:50 |
| *** derekh has quit IRC | 11:50 | |
| janders | to wrap up my FC32 bifrost install blues - I got it running in the end. Needed ignore_errors in all ironic and inspector DB tasks and a couple re-runs. I have no reason to think those are coding issues though, it seems the only change needed is removing libselinux-python. | 11:52 |
| janders | will re-test on better hardware later in the week hopefully and report back. I'm happy to submit that one-line change, it will be good contributing practice for me :) | 11:54 |
| janders | thanks for all your help and will keep you posted | 11:54 |
| *** priteau has joined #openstack-ironic | 11:55 | |
| *** yaawang has quit IRC | 11:58 | |
| *** yaawang has joined #openstack-ironic | 11:59 | |
| *** janders has quit IRC | 12:01 | |
| *** rh-jelabarre has joined #openstack-ironic | 12:02 | |
| *** priteau has quit IRC | 12:12 | |
| dtantsur | \o/ | 12:15 |
| *** Qianbiao has quit IRC | 12:20 | |
| *** priteau has joined #openstack-ironic | 12:20 | |
| maelk | Hi! Do you think the underlying error in https://github.com/metal3-io/baremetal-operator/issues/578 would need to be fixed in Ironic side ? basically, when IPA is booted out of turn, i.e. node is already introspected and powered off by ironic, but then powered on by BMO in this case and restrict_lookup is true, it somehow starts again an introspect | 12:22 |
| maelk | ion, reports to inspector and makes it throw an exception, and loop until another operation is performed from ironic side. Is this worth a story ? or do you consider that the node should not boot IPA anyways ? | 12:22 |
| rpittau | maelk: just my opinion but in theory it should not boot in IPA again | 12:29 |
| *** Qianbiao has joined #openstack-ironic | 12:29 | |
| * rpittau is the king of typos | 12:30 | |
| *** diurnalist has joined #openstack-ironic | 12:32 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 12:33 |
| dtantsur | iurygregory: could you approve https://review.opendev.org/#/c/739244/ please? | 12:34 |
| patchbot | patch 739244 - ironic - Follow up of fix uefi jobs with ovmf native ubuntu... - 2 patch sets | 12:34 |
| iurygregory | dtantsur, sure | 12:35 |
| *** diurnalist has quit IRC | 12:37 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/ussuri: Increase vm ram size in uefi netoobted jobs https://review.opendev.org/739728 | 12:39 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/ussuri: Increase vm ram size in uefi netbooted jobs https://review.opendev.org/739728 | 12:41 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: Use node.id instead of node.uuid in record_node https://review.opendev.org/739730 | 12:41 |
| dtantsur | rpittau: ^^ | 12:41 |
| dtantsur | maelk: what would be the expected behavior? | 12:42 |
| maelk | that inspector would not throw an exception at least, and either discard the input ( to avoid some kind of spoofing maybe) or update it if it is changed (but how to make sure that it is not spoofed ? ) | 12:44 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Stop using md5 for __repr__ of objects https://review.opendev.org/739735 | 12:53 |
| arne_wiebalck | dtantsur: maelk: why does the node start another inspection after booting into the IPA? The state in the cache should be finished, no? | 12:59 |
| dtantsur | arne_wiebalck: it doesn't, it fails because of a bug in the code that checks for active node introspection | 13:00 |
| dtantsur | see the patch above | 13:00 |
| arne_wiebalck | dtantsur: ah, ok, sorry, I missed this | 13:00 |
| rpittau | that's probably a leftover from the openstacksdk migration | 13:00 |
| dtantsur | it is | 13:01 |
| maelk | we can see that the IPA reports its introspection data after being booted | 13:01 |
| maelk | when calling /v1/continue | 13:01 |
| dtantsur | that's expected, IPA has no way to know that it's not supposed to run | 13:01 |
| maelk | ok, good | 13:02 |
| TheJulia | good morning | 13:02 |
| rpittau | good morning TheJulia :) | 13:03 |
| dtantsur | TheJulia: morning! | 13:05 |
| maelk | dtantsur thank you for the fix. | 13:05 |
| dtantsur | will you believe if I tell you that the CI got broken on the release day? | 13:05 |
| iurygregory | it's normal for us | 13:08 |
| TheJulia | dtantsur: seriously?? | 13:08 |
| iurygregory | not the 1st time =X | 13:08 |
| * TheJulia sighs | 13:08 | |
| *** diurnalist has joined #openstack-ironic | 13:08 | |
| iurygregory | good morning TheJulia =) | 13:09 |
| TheJulia | btw: re fips, there is some chatter out there about crc32 needing to be replaced with sha referecing the fips 140-3 standard, however the standard just says "go look at these documents" | 13:09 |
| iurygregory | facepalm | 13:12 |
| dtantsur | TheJulia: I hope it's not shutting down usages of crc32, otherwise anything using ZIP and many other algos is screwed | 13:13 |
| dtantsur | and yes, we have a CI problem. the fix has been approved already. | 13:13 |
| *** diurnalist has quit IRC | 13:13 | |
| TheJulia | dtantsur: the comments were it is, but I'll keep digging once I have coffee | 13:18 |
| *** Goneri has joined #openstack-ironic | 13:18 | |
| TheJulia | arne_wiebalck: replied to your comment on https://review.opendev.org/#/c/726483/5 I think you missed the maximum loop count. :) | 13:18 |
| patchbot | patch 726483 - ironic - Guard conductor from consuming all of the ram - 5 patch sets | 13:18 |
| *** derekh has joined #openstack-ironic | 13:19 | |
| arne_wiebalck | TheJulia: The operation will time out, but the memory will never become free again I think. | 13:19 |
| *** moguimar has joined #openstack-ironic | 13:20 | |
| TheJulia | did you see my comment as to what memory_available is? | 13:20 |
| * TheJulia goes and grabs coffee | 13:20 | |
| dtantsur | crc32 is not allowed to be used for cryptography, but if it's disabled completely.. many things won't work | 13:21 |
| *** moguimar has left #openstack-ironic | 13:21 | |
| arne_wiebalck | From what I see (and tested), the available memory is what is available immediately (so not buffered/cached). Won't this go down over time as caches fill up? | 13:21 |
| *** rloo has joined #openstack-ironic | 13:22 | |
| dtantsur | but honestly, there is no reason to use even crc32 in this patch | 13:22 |
| TheJulia | hmm, that is problematic then | 13:23 |
| TheJulia | that was for arne_wiebalck | 13:23 |
| TheJulia | arne_wiebalck: I guess at worst we could always look at available + cache | 13:24 |
| arne_wiebalck | TheJulia: maybe I am mixing up definitions and understanding | 13:26 |
| arne_wiebalck | TheJulia: I was just wondering if it is enough to wait for RAM to become free or whether we should consider to drop caches regularly | 13:27 |
| *** Qianbiao has quit IRC | 13:27 | |
| *** Qianbiao has joined #openstack-ironic | 13:27 | |
| arne_wiebalck | TheJulia: I think you are right: available includes the caches which could be dropped. So dropping caches will not change available. | 13:28 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Stop using md5 for __repr__ of objects https://review.opendev.org/739735 | 13:28 |
| dtantsur | even simpler ^^ | 13:28 |
| arne_wiebalck | TheJulia: 'free' should drop when caches are flushed | 13:29 |
| arne_wiebalck | TheJulia: yes, just confirmed this on an Ironic controller: 'available' does not change when the caches are dropped | 13:32 |
| *** cdearborn has joined #openstack-ironic | 13:32 | |
| *** Goneri has quit IRC | 13:33 | |
| TheJulia | dtantsur: FIPS 140-3 is super vague. https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-140C.pdf is what is only permitted in the context/confines of security. However that too is kind of vague. CRC may fall into a bit of a gray area | 13:33 |
| TheJulia | arne_wiebalck: we shouldn't try to drive the cache to drop, that is super bad and technically you have to touch the tunables to suggest to the kernel to do it | 13:34 |
| TheJulia | arne_wiebalck: simple enough to change | 13:34 |
| arne_wiebalck | TheJulia: I think we are good with available, I mixed this with free (which will go down with time) | 13:35 |
| TheJulia | arne_wiebalck: but I wonder if we should be using free | 13:36 |
| TheJulia | yay for words! | 13:36 |
| *** ricolin has joined #openstack-ironic | 13:36 | |
| arne_wiebalck | TheJulia: If we were to use free, we would need to free up memory ourselves. | 13:37 |
| dtantsur | TheJulia: what can be better than a vague guideline for security? :) anyway, I've updated the patch to not use crc32 | 13:37 |
| TheJulia | arne_wiebalck: "Do not set this too high!" warning sounds in order | 13:38 |
| arne_wiebalck | TheJulia: At least this is my understanding, since the kernel will make use of all memory over time, for caching. So free will become very small. I think available is the correct metric. | 13:38 |
| arne_wiebalck | TheJulia: Did you test this with real hardware? | 13:38 |
| TheJulia | memory exhaustion, no but I'm poking at my desktop's memory freeness now | 13:39 |
| *** Qianbiao has quit IRC | 13:39 | |
| TheJulia | I can load it up on the nuc and and see if it complains if I slowly turn down the amount of available free memory | 13:39 |
| *** Qianbiao has joined #openstack-ironic | 13:39 | |
| TheJulia | my desktop currently reports 13 GB free, 17 GB available (which is surprising, tbh) | 13:40 |
| TheJulia | yeah, and the difference is if the buffer/cache is released | 13:41 |
| openstackgerrit | Merged openstack/bifrost stable/ussuri: Do not use 'sudo pip install' when venv is used https://review.opendev.org/739718 | 13:41 |
| arne_wiebalck | which the kernel will do if the conductor asks for it | 13:41 |
| TheJulia | yup | 13:41 |
| arne_wiebalck | I think we're good, sorry for the noise. | 13:42 |
| TheJulia | "Please tune wisely" note seems super applicable though | 13:42 |
| arne_wiebalck | Yes. In particular in view of the RSS footprint kaifeng and I have seen/discussed. So, if the conductor is run alongside the API, we may run out of steam quite quickly. | 13:45 |
| TheJulia | can we put a picture of a light colored unix beard in the documents suggesting "consult with the unix bearded one" before making arbitrary memory setting changes | 13:46 |
| arne_wiebalck | :-D | 13:46 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Follow up of enforce autospec in test_manager https://review.opendev.org/739063 | 13:46 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Enforce autospec in test_task_manager https://review.opendev.org/739758 | 13:46 |
| TheJulia | "consult your available unix beards or grumpy unix ladies before tuning memory" | 13:47 |
| TheJulia | arne_wiebalck: are there memory issues in the API? | 13:47 |
| arne_wiebalck | TheJulia: the footprint is surprisingly big, yes | 13:48 |
| TheJulia | hmm | 13:48 |
| arne_wiebalck | TheJulia: we observe RSS a few times the size of the entire DB | 13:48 |
| * TheJulia blinks | 13:48 | |
| TheJulia | db + indexes, or raw data in the bd? | 13:49 |
| TheJulia | db | 13:49 |
| arne_wiebalck | TheJulia: this can be influenced with the processes/workers | 13:49 |
| *** tzumainn has joined #openstack-ironic | 13:49 | |
| TheJulia | oh yeah | 13:49 |
| TheJulia | each process will cary a copy of the hash ring | 13:49 |
| TheJulia | carry | 13:49 |
| arne_wiebalck | TheJulia: well, a dump of our DB is around 60MB or so | 13:49 |
| arne_wiebalck | TheJulia: somewhere I have notes about process and threads per process along with their RSS footprint | 13:50 |
| TheJulia | I think the last time I looked we were at like ~230mb with an empty database | 13:50 |
| arne_wiebalck | Yeah, that sounds familiar. | 13:51 |
| arne_wiebalck | But with a few processes and threads this multiplies quite quickly | 13:51 |
| arne_wiebalck | And when you then start image conversion ... | 13:52 |
| arne_wiebalck | I have not split the API from the conductors, they run in pairs on each host. | 13:52 |
| arne_wiebalck | And the inspector is there as well, to make the Ironic party complete and exclude noone. | 13:54 |
| TheJulia | Yeah, so I could see you hitting memory issues super easily in that case with increased load because depending on how your running webserver processes you may find yourself with more being spawned | 13:54 |
| arne_wiebalck | Yeah, maybe it is time to split the services. | 13:56 |
| TheJulia | I would say that depends on your burst load :) | 13:58 |
| *** Goneri has joined #openstack-ironic | 14:02 | |
| arne_wiebalck | it is bursty, calm for weeks/months, then hundreds of nodes in quick succession | 14:02 |
| arne_wiebalck | this is why I have the "wing" conductors | 14:02 |
| arne_wiebalck | the leading conductor group has more conductors than the other ones | 14:02 |
| arne_wiebalck | since the activity is in this group | 14:03 |
| arne_wiebalck | and I planned to add two more upon delivery arrival | 14:04 |
| arne_wiebalck | since they are all equal and on VMs, this is pretty quick and easy | 14:04 |
| *** abdysn has quit IRC | 14:20 | |
| openstackgerrit | Merged openstack/ironic master: Follow up of fix uefi jobs with ovmf native ubuntu package https://review.opendev.org/739244 | 14:32 |
| rpittau | I can go on vacation now | 14:33 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Use assertCountEqual instead of assertItemsEqual, part 2 https://review.opendev.org/739776 | 14:33 |
| rpittau | ok, after this ^ | 14:34 |
| iurygregory | rpittau, enjoy! =) | 14:35 |
| rpittau | eheh | 14:36 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Add an option to choose the hash ring algorithm https://review.opendev.org/739779 | 14:39 |
| *** diurnalist has joined #openstack-ironic | 14:47 | |
| TheJulia | arne_wiebalck: I'm typing up a quick doc on tuning, if you have thoughts feel free to add to it but hopefully it gets a lot of it out of my head | 14:48 |
| arne_wiebalck | TheJulia: +1 | 14:49 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Add some tuning documentation https://review.opendev.org/739783 | 14:49 |
| *** hjensas has quit IRC | 14:50 | |
| *** hjensas has joined #openstack-ironic | 14:52 | |
| * dtantsur wonders why his server machine has firefox installed | 14:53 | |
| iurygregory | O.o | 14:58 |
| iurygregory | wow | 14:58 |
| TheJulia | dtantsur: anything install selenium? | 14:58 |
| dtantsur | that's a possibility | 14:58 |
| * dtantsur starts a deployment that is doomed to fail | 14:58 | |
| iurygregory | =( | 14:59 |
| dtantsur | mmmm, I did not expect it to fail SO quickly | 15:02 |
| iurygregory | enr bifrost failed on F32 [bifrost-ironic-install : Create ironic DB Schema] | 15:06 |
| *** k_mouza_ has joined #openstack-ironic | 15:11 | |
| dtantsur | ouch? so there is actually a problem with mysql? | 15:12 |
| *** k_mouza has quit IRC | 15:13 | |
| iurygregory | seems to be .-. | 15:17 |
| iurygregory | I had no problem 1month ago... | 15:17 |
| TheJulia | iurygregory: can you put the errors in a paste? | 15:21 |
| iurygregory | http://paste.openstack.org/show/795623/ | 15:22 |
| TheJulia | iurygregory: do you get any errors if you manually try to create the schema? | 15:23 |
| iurygregory | TheJulia, I was about to try this =) | 15:23 |
| TheJulia | it seems SUPER weird that you have no output | 15:25 |
| iurygregory | yeah | 15:25 |
| iurygregory | doesn't help =( | 15:25 |
| TheJulia | does anything get written to the journal or syslog facility? | 15:26 |
| dtantsur | TheJulia: janders had exactly the same problem today | 15:26 |
| prometheanfire | dtantsur: rpittau: looks like it might be a new mock issue then? https://ba29496b3bf7c2dae4e1-b2ee54f1bb7128298daf0c74e8f5e0db.ssl.cf2.rackcdn.com/739652/1/check/cross-ironic-py38/f8eea70/testr_results.html | 15:28 |
| iurygregory | ouch! | 15:29 |
| iurygregory | this doesn't look good | 15:29 |
| dtantsur | prometheanfire: that's the price we pay for verifying error messages from a 3rd party library | 15:29 |
| iurygregory | http://paste.openstack.org/show/795626/ | 15:29 |
| rpittau | ok, that looks like jsonpatch | 15:29 |
| iurygregory | probably py38 problems... | 15:29 |
| TheJulia | iurygregory: that is chilling | 15:30 |
| dtantsur | should be a simple fix, anyone doing it or should I? | 15:30 |
| prometheanfire | dtantsur: :D | 15:30 |
| rpittau | dtantsur: go ahead :) | 15:30 |
| iurygregory | dtantsur, you talking about bifrost or about the mock issue? | 15:31 |
| dtantsur | iurygregory: the unit test issue | 15:31 |
| dtantsur | (it's not related to mock any more) | 15:31 |
| iurygregory | I can take care =) | 15:31 |
| iurygregory | if you don't mind ofc | 15:31 |
| TheJulia | iurygregory: is mysqld-safe running after the failure? Has it logged anything? | 15:32 |
| iurygregory | TheJulia, let me check | 15:33 |
| * TheJulia really should get checked for adult ADD | 15:35 | |
| TheJulia | or I should just juggle less | 15:35 |
| iurygregory | TheJulia, couldn't find anything for mysqld-safe, I've found some mariadb logs http://paste.openstack.org/show/795627/ | 15:38 |
| TheJulia | wow | 15:40 |
| TheJulia | that is bad | 15:40 |
| TheJulia | is this a brand new mariadb rpm? | 15:40 |
| iurygregory | mysql Ver 15.1 Distrib 10.4.13-MariaDB, for Linux (x86_64) | 15:41 |
| iurygregory | the version that got installed in F32 | 15:41 |
| iurygregory | let me check the rpm info | 15:41 |
| TheJulia | iurygregory: also, anything in the kernel log? | 15:41 |
| TheJulia | the kernel often doesn't issue sig11 on it's own... | 15:41 |
| iurygregory | will check | 15:41 |
| *** ociuhandu_ has joined #openstack-ironic | 15:43 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Do not verify error messages from jsonpatch in unit tests https://review.opendev.org/739803 | 15:44 |
| dtantsur | prometheanfire: ^^ | 15:44 |
| prometheanfire | cool :D | 15:46 |
| *** ociuhandu has quit IRC | 15:46 | |
| *** ociuhandu_ has quit IRC | 15:47 | |
| rpittau | well, at least we know that the cross-test works :D | 15:47 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Provide a path to set explicit ipxe bootloaders https://review.opendev.org/728315 | 15:47 |
| prometheanfire | rpittau: true, it has proved useful | 15:48 |
| *** dtantsur is now known as dtantsur|brb | 15:49 | |
| iurygregory | strange, it seems that the version installed is not even listed in https://fedora.pkgs.org/32/fedora-x86_64/mariadb-server-10.4.12-3.fc32.x86_64.rpm.html O.o | 15:49 |
| *** tzumainn has quit IRC | 15:50 | |
| iurygregory | mariadb-server-10.4.13-2.fc32.x86_64 O.o | 15:50 |
| rpittau | iurygregory: that's kind of weird, let me check on a clean F32 machine | 15:51 |
| TheJulia | ++ | 15:51 |
| openstackgerrit | Merged openstack/ironic-inspector master: Use node.id instead of node.uuid in record_node https://review.opendev.org/739730 | 15:51 |
| iurygregory | rpittau, well this was on a clean F32... =( | 15:51 |
| TheJulia | err | 15:51 |
| iurygregory | .-. | 15:52 |
| TheJulia | is this the first time on F32, or is this a brand new thing on FC32 + updates | 15:52 |
| *** tzumainn has joined #openstack-ironic | 15:52 | |
| rpittau | iurygregory: did you run dnf ugprade after teh install ? | 15:52 |
| iurygregory | rpittau, nope, only dnf install git, dnf install @virtualization | 15:53 |
| *** Qianbiao has quit IRC | 15:55 | |
| TheJulia | Did the upgrade still get pulled in? | 15:55 |
| iurygregory | maybe they change the image I've tired (1month ago) | 15:56 |
| TheJulia | Where was this image from? | 15:56 |
| iurygregory | internal repos | 15:57 |
| TheJulia | rpittau: pulling down public? | 15:59 |
| rpittau | I'm checking the version locally and also in one of my testing vms, I have mariadb-server-3:10.4.13-2.fc32.x86_64 | 16:00 |
| rpittau | so I guess a rebuild is needed there | 16:00 |
| rpittau | plus a dnf ugprade | 16:00 |
| rpittau | or upgrade is better :) | 16:01 |
| * rpittau particularly dyslexic today | 16:01 | |
| *** lucasagomes has quit IRC | 16:01 | |
| TheJulia | iurygregory: nothing in the kernel log? | 16:04 |
| * TheJulia hopes for a double free() message | 16:05 | |
| iurygregory | TheJulia, sorry >.< let me copy to a paste | 16:05 |
| TheJulia | no worries | 16:05 |
| *** tzumainn has quit IRC | 16:05 | |
| rpittau | I confirm that the installer is fetching mariadb-server-3:10.4.13-2.fc32.x86_64 | 16:07 |
| rpittau | I'm running the bifrost installer playbook now | 16:07 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 16:10 |
| TheJulia | rpittau: is it getting glibc at all? | 16:12 |
| iurygregory | dmesg http://paste.openstack.org/show/795630/ | 16:13 |
| TheJulia | iurygregory: looks like it got truncated? | 16:14 |
| iurygregory | yeah =( | 16:14 |
| iurygregory | let me see if I can do something about it | 16:14 |
| iurygregory | I can give you access to the machine TheJulia if you have a key | 16:15 |
| TheJulia | iurygregory: sending | 16:15 |
| *** ricolin has quit IRC | 16:17 | |
| * iurygregory needs to prepare a fast dinner (meeting in a few minutes) | 16:18 | |
| TheJulia | heh, it has a wifi adapter | 16:20 |
| TheJulia | looks like a minor selinux issue but it still looks permissive | 16:21 |
| TheJulia | nothing in kernel log to indicate the kernel killed it :\ | 16:22 |
| TheJulia | so I guess it must have internallyf aulted | 16:22 |
| TheJulia | faulted | 16:22 |
| *** ociuhandu has joined #openstack-ironic | 16:22 | |
| iurygregory | =( | 16:22 |
| TheJulia | I guess we can wait for rpittau to report back what he encounters. Hopefully we have enough info to make a cogent bug | 16:26 |
| *** ociuhandu has quit IRC | 16:26 | |
| *** tzumainn has joined #openstack-ironic | 16:28 | |
| * rpittau reports back | 16:29 | |
| rpittau | it works on my vm :) | 16:29 |
| TheJulia | is selinux disabled or enabled? | 16:30 |
| rpittau | Enforcing | 16:30 |
| TheJulia | rpittau: did you already have an ironic database? | 16:31 |
| TheJulia | iurygregory: same question^ | 16:31 |
| rpittau | no, this is a fresh new F32 machine, I ran dnf upgrade, rebooted and then ran the bifrost installer | 16:31 |
| iurygregory | TheJulia, nope it was a fresh env | 16:32 |
| * iurygregory didn't run dnf upgrade... | 16:32 | |
| * TheJulia goes and reads the discussion with janders | 16:32 | |
| rpittau | I just modified the selinux package cause as we know it throws an error | 16:32 |
| TheJulia | jacob said it was intermittent | 16:34 |
| TheJulia | iurygregory: just to double confirm, the 10.4.13-2 rpm was what was installed? | 16:37 |
| iurygregory | TheJulia, correct | 16:37 |
| TheJulia | looks like the main rpm was updated, but the "updates" repo has 10.4.13-2 as the latest | 16:39 |
| rpittau | I will run it again | 16:40 |
| TheJulia | The changelog shows 10.4.13-2 was last updated, at least before the rpm was built on 6/5/2020 | 16:41 |
| TheJulia | 5th of last month | 16:41 |
| *** dtantsur|brb is now known as dtantsur | 16:41 | |
| dtantsur | TheJulia: Jacob did manage to get a working environment by inserting a lot of ignore_errors and retrying many times | 16:42 |
| *** SpamapS has quit IRC | 16:42 | |
| TheJulia | yeah | 16:43 |
| TheJulia | iurygregory: could you try putting the latest 10.4.13-3 rpm on there and giving it a try | 16:43 |
| TheJulia | The C client library connector code was rebased to fix https://jira.mariadb.org/browse/CONC-441 | 16:44 |
| TheJulia | while seemingly unrelated, I'm wondering if things maically work | 16:45 |
| TheJulia | since effective and db user used can differ | 16:45 |
| dtantsur | rpittau: cannot figure out from release notes, when was the switch to openstacksdk for inspector: ussuri or victoria? | 16:45 |
| rpittau | ussuri | 16:45 |
| dtantsur | k thx | 16:46 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector stable/ussuri: Use node.id instead of node.uuid in record_node https://review.opendev.org/739813 | 16:46 |
| rpittau | dtantsur: https://opendev.org/openstack/ironic-inspector/commit/3accdfbbc6cf7176f9ff8b4967195cb4e2036cea | 16:46 |
| dtantsur | yeah | 16:46 |
| *** Lucas_Gray has quit IRC | 16:47 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: Add release version to release notes https://review.opendev.org/739814 | 16:48 |
| *** diurnalist has quit IRC | 16:50 | |
| rpittau | ok so now I'm seeing the error | 16:51 |
| rpittau | exactly the same one | 16:52 |
| TheJulia | sounds like it is time for a very verbose BZ | 16:54 |
| rpittau | it crashes and recover | 16:56 |
| rpittau | this is not nice | 16:56 |
| rpittau | [ 2712.243361] mysqld[17777]: segfault at 8 ip 00005632a0d0706b sp 00007fb7100d23b0 error 4 in mysqld[5632a087f000+8b9000] | 16:56 |
| dtantsur | ugh | 16:56 |
| rpittau | TheJulia: do you want the stack trace from journalctl ? | 16:57 |
| TheJulia | well, at least your machine logged it | 16:57 |
| TheJulia | rpittau: is it the same as what iurygregory got? | 16:57 |
| iurygregory | TheJulia, will give a try | 16:58 |
| iurygregory | I was having dinner | 16:58 |
| TheJulia | oh no worries | 16:58 |
| rpittau | oh, not sure, this is from systemd-coredump | 16:58 |
| rpittau | http://paste.openstack.org/show/795634/ | 16:59 |
| *** diurnalist has joined #openstack-ironic | 17:02 | |
| *** sziviani has quit IRC | 17:19 | |
| *** sziviani has joined #openstack-ironic | 17:20 | |
| *** SpamapS has joined #openstack-ironic | 17:20 | |
| *** derekh has quit IRC | 17:27 | |
| *** dtantsur is now known as dtantsur|afk | 17:38 | |
| dtantsur|afk | g'night! | 17:38 |
| TheJulia | o/ | 17:40 |
| TheJulia | rpittau: that is no good :( | 17:41 |
| TheJulia | but it aligns | 17:41 |
| TheJulia | anyone already started on a bugzilla? | 17:41 |
| *** k_mouza_ has quit IRC | 17:49 | |
| *** priteau has quit IRC | 17:52 | |
| arne_wiebalck | Bye everyone o/ | 17:52 |
| openstackgerrit | Merged openstack/ironic master: Use assertCountEqual instead of assertItemsEqual, part 2 https://review.opendev.org/739776 | 17:56 |
| iurygregory | seems like ironic-tempest-ipa-partition-pxe_ipmitool didn't like basic_ops running with partition/wholedisk .-. | 17:56 |
| TheJulia | iurygregory: huh? | 18:31 |
| *** ociuhandu has joined #openstack-ironic | 18:40 | |
| *** ociuhandu has quit IRC | 18:45 | |
| *** aedc has joined #openstack-ironic | 19:04 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Provide a path to set explicit ipxe bootloaders https://review.opendev.org/728315 | 19:35 |
| TheJulia | gah! | 19:35 |
| TheJulia | screwed up | 19:35 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Provide a path to set explicit ipxe bootloaders https://review.opendev.org/728315 | 19:38 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Add ``ipxe_template`` override option https://review.opendev.org/739841 | 19:44 |
| TheJulia | much better | 19:44 |
| iurygregory | TheJulia, yeah, I'm going to try to reproduce locally tomorrow morning but seems that ironic-tempest-ipa-partition-pxe_ipmitool is the only job that is unhappy with basic_ops for wholedisk (since I have the change that adds partition/wholedisk test to basic_ops) https://review.opendev.org/735960 https://review.opendev.org/714453 | 19:52 |
| patchbot | patch 735960 - ironic-tempest-plugin - Improve test_baremetal_basic_ops - 5 patch sets | 19:52 |
| patchbot | patch 714453 - ironic - add tempest boot_mode config - 17 patch sets | 19:52 |
| TheJulia | Okay, have a wonderful evening | 20:02 |
| *** Lucas_Gray has joined #openstack-ironic | 20:31 | |
| openstackgerrit | Merged openstack/ironic master: Do not put the whole JSON error from the agent in last_error https://review.opendev.org/739439 | 20:50 |
| openstackgerrit | Merged openstack/ironic master: Fix missing print format in log message https://review.opendev.org/739667 | 21:05 |
| *** hjensas has quit IRC | 21:21 | |
| *** hjensas has joined #openstack-ironic | 21:52 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Add some tuning documentation https://review.opendev.org/739783 | 21:52 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Add ``ipxe_template`` override option https://review.opendev.org/739841 | 22:00 |
| openstackgerrit | Merged openstack/ironic-inspector stable/ussuri: Use node.id instead of node.uuid in record_node https://review.opendev.org/739813 | 22:17 |
| *** dsneddon has joined #openstack-ironic | 22:25 | |
| *** rloo has quit IRC | 22:26 | |
| *** jungleboyj has quit IRC | 22:43 | |
| *** samueldmq has quit IRC | 22:44 | |
| *** erbarr has quit IRC | 22:45 | |
| *** jungleboyj has joined #openstack-ironic | 22:45 | |
| *** antonym has joined #openstack-ironic | 22:46 | |
| *** seongsoocho has quit IRC | 22:47 | |
| *** gmann has quit IRC | 22:47 | |
| *** vkmc has quit IRC | 22:47 | |
| *** m1p has quit IRC | 22:47 | |
| *** phrobb has quit IRC | 22:48 | |
| *** erbarr has joined #openstack-ironic | 22:48 | |
| *** samueldmq has joined #openstack-ironic | 22:48 | |
| *** tonyb[m] has quit IRC | 22:50 | |
| *** mnaser has quit IRC | 22:50 | |
| *** vdrok has quit IRC | 22:50 | |
| *** TheJulia has quit IRC | 22:50 | |
| *** rpioso has quit IRC | 22:50 | |
| *** buhman has quit IRC | 22:50 | |
| *** portdirect has quit IRC | 22:51 | |
| *** ildikov has quit IRC | 22:51 | |
| *** johnsom has quit IRC | 22:51 | |
| *** andrein has quit IRC | 22:51 | |
| *** johnhaan has quit IRC | 22:51 | |
| *** rpittau has quit IRC | 22:51 | |
| *** pas-ha has quit IRC | 22:52 | |
| *** arne_wiebalck has quit IRC | 22:52 | |
| *** NobodyCam has quit IRC | 22:52 | |
| *** cdearborn has quit IRC | 22:52 | |
| *** erbarr has quit IRC | 22:52 | |
| *** rajinir_ has quit IRC | 22:52 | |
| *** samueldmq has quit IRC | 22:53 | |
| *** jungleboyj has quit IRC | 22:53 | |
| *** v1k0d3n has quit IRC | 22:53 | |
| *** donnyd has quit IRC | 22:53 | |
| *** tkajinam has joined #openstack-ironic | 22:54 | |
| *** rcernin has joined #openstack-ironic | 23:17 | |
| *** Goneri has quit IRC | 23:25 | |
| *** Lucas_Gray has quit IRC | 23:51 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!