| *** lbotell1 has quit IRC | 00:00 | |
| *** rmcall has quit IRC | 00:06 | |
| TheJulia | stevebaker: much appreciated | 00:24 |
|---|---|---|
| TheJulia | the bad gateway one where it seems apache blows up occassionally on missing content type... seems worthwhile to go ahead and fix since we're getting something like 1 out of every 5-10 runs of that job as that failure right now | 00:24 |
| *** gyee has quit IRC | 00:36 | |
| *** k_mouza has joined #openstack-ironic | 00:51 | |
| *** k_mouza has quit IRC | 00:56 | |
| *** k_mouza has joined #openstack-ironic | 00:58 | |
| *** ricolin_ has joined #openstack-ironic | 00:59 | |
| *** k_mouza has quit IRC | 01:04 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Decompose the core deploy step on iscsi and ansible deploy https://review.opendev.org/731180 | 01:07 |
| *** uzumaki has joined #openstack-ironic | 01:10 | |
| *** janders has joined #openstack-ironic | 01:21 | |
| *** k_mouza has joined #openstack-ironic | 01:27 | |
| *** k_mouza has quit IRC | 01:31 | |
| *** Goneri has quit IRC | 01:43 | |
| uzumaki | I just did a bit of bug triage.. I think.. I'm unsure where to list those stories on the whiteboard? the doc said for the "RFE Review" section but I dont' find any.. also, found a couple stories that can be closed because of being stale | 02:14 |
| uzumaki | I've triaged around 16 stories, some simply needed tags (so they don't show up on the stats anymore as untriaged) | 02:15 |
| *** rmcall has joined #openstack-ironic | 02:31 | |
| *** tzumainn has quit IRC | 02:46 | |
| *** rcernin has quit IRC | 02:56 | |
| *** mkrai has joined #openstack-ironic | 02:57 | |
| *** rcernin has joined #openstack-ironic | 02:57 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Allow deleting nodes with a broken driver https://review.opendev.org/739702 | 02:58 |
| *** ociuhandu has joined #openstack-ironic | 03:01 | |
| *** ociuhandu has quit IRC | 03:06 | |
| *** k_mouza has joined #openstack-ironic | 03:44 | |
| *** k_mouza has quit IRC | 03:48 | |
| openstackgerrit | Steve Baker proposed openstack/ironic-inspector master: Fix hacking min version to 3.0.1 https://review.opendev.org/730177 | 03:56 |
| *** yaawang has quit IRC | 04:10 | |
| *** yaawang has joined #openstack-ironic | 04:11 | |
| *** bfournie has quit IRC | 04:11 | |
| *** bfournie has joined #openstack-ironic | 04:12 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Do not validate driver on changing non-driver fields https://review.opendev.org/739699 | 04:32 |
| openstackgerrit | Verification of a change to openstack/ironic-inspector failed: Make inspector base CI job use 2 CPUs https://review.opendev.org/740746 | 04:42 |
| *** diurnalist has quit IRC | 05:13 | |
| *** k_mouza has joined #openstack-ironic | 05:13 | |
| *** yaawang has quit IRC | 05:16 | |
| *** abdysn has joined #openstack-ironic | 05:16 | |
| *** yaawang has joined #openstack-ironic | 05:16 | |
| *** k_mouza has quit IRC | 05:18 | |
| *** diurnalist has joined #openstack-ironic | 05:49 | |
| *** diurnalist has quit IRC | 05:53 | |
| *** diurnalist has joined #openstack-ironic | 05:59 | |
| *** diurnalist has quit IRC | 06:04 | |
| *** uzumaki has quit IRC | 06:18 | |
| *** yaawang has quit IRC | 06:26 | |
| *** yaawang has joined #openstack-ironic | 06:27 | |
| arne_wiebalck | Good morning, ironic! | 06:34 |
| arne_wiebalck | TheJulia: We will get some new hardware in a couple of weeks and the amount of servers should be enough to test the OOM guard patch ... I will need to see if I can prepare things to squeeze testing in as we need these servers in prod as well :) | 06:36 |
| *** diurnalist has joined #openstack-ironic | 06:41 | |
| *** diurnalist has quit IRC | 06:45 | |
| janders | g'day Arne o/ | 06:56 |
| iurygregory | good morning arne_wiebalck and Ironic! | 06:56 |
| iurygregory | hey janders o/ | 06:56 |
| janders | good morning iurygregory | 06:56 |
| janders | o/ | 06:56 |
| arne_wiebalck | hey iurygregory & janders o/ | 06:57 |
| iurygregory | arne_wiebalck, any thoughts on https://review.opendev.org/#/c/740309/ ? I think it covers your idea to keep the devices with 0 for cleanning | 07:01 |
| patchbot | patch 740309 - ironic-python-agent - Ignore devices with size 0 when collecting inventory - 3 patch sets | 07:01 |
| *** dsneddon has quit IRC | 07:02 | |
| arne_wiebalck | iurygregory: will have a look, but have 3 meetings lined up this morning ... | 07:02 |
| iurygregory | arne_wiebalck, ack | 07:02 |
| iurygregory | good luck! | 07:02 |
| *** rcernin has quit IRC | 07:07 | |
| *** k_mouza has joined #openstack-ironic | 07:09 | |
| *** mkrai has quit IRC | 07:13 | |
| *** k_mouza has quit IRC | 07:13 | |
| *** ociuhandu has joined #openstack-ironic | 07:17 | |
| *** rcernin has joined #openstack-ironic | 07:17 | |
| *** k_mouza has joined #openstack-ironic | 07:18 | |
| *** k_mouza has quit IRC | 07:22 | |
| *** ociuhandu_ has joined #openstack-ironic | 07:28 | |
| *** ociuhandu has quit IRC | 07:31 | |
| rpittau | good morning ironic o/ | 07:43 |
| *** dtantsur|afk is now known as dtantsur | 07:45 | |
| dtantsur | morning ironic | 07:45 |
| iurygregory | morning rpittau and dtantsur o/ | 07:46 |
| rpittau | hey iurygregory :) | 07:46 |
| iurygregory | dtantsur, do we want to switch ironic-lib to tinyipa or nope? | 07:46 |
| iurygregory | rpittau, I hope you enjoyed the PTO =) | 07:47 |
| dtantsur | dhellmann: instance_info/deploy_boot_mode is what you need; updating properties/capabilities does involve messing with strings | 07:47 |
| rpittau | I did, although it was not exactly "relaxing" | 07:47 |
| dtantsur | iurygregory: please don't recommend people to put boot_mode to instance_info/capabilities, the last time I checked it didn't work | 07:48 |
| dtantsur | (the only capability that does NOT work this way, because... consistency?) | 07:48 |
| iurygregory | =O | 07:48 |
| dtantsur | iurygregory: ironic-lib - I'm afraid so :( | 07:48 |
| dtantsur | are we still hit by the EPEL problem? | 07:48 |
| rpittau | still issues with EPEL ? | 07:48 |
| * rpittau dives deep into emails | 07:49 | |
| iurygregory | \o/ oh we probably don't need according to the new logs | 07:49 |
| iurygregory | but we do need 2 VM's =) | 07:50 |
| * iurygregory sends the patch | 07:50 | |
| janders | good morning dtantsur rpittau o/ | 07:53 |
| rpittau | hey janders :) | 07:53 |
| dtantsur | iurygregory: 2 VMs and only 2 jobs, right? | 07:53 |
| dtantsur | as with IPA? | 07:53 |
| *** k_mouza has joined #openstack-ironic | 07:54 | |
| janders | dtantsur: are you happy with me picking up https://storyboard.openstack.org/#!/story/1618014? iurygregory helpfully pointed out that https://storyboard.openstack.org/#!/story/2007646 which I was looking at earlier isn't rfe-approved | 07:56 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic-lib master: Update ironic-lib jobs https://review.opendev.org/741105 | 07:56 |
| iurygregory | dtantsur, yeah | 07:56 |
| iurygregory | the fix ^ | 07:56 |
| iurygregory | and I've removed 2 jobs \o/ | 07:56 |
| dtantsur | janders: I *think* the 2nd one was not approved by mistake. if you check the comments, at least the 1st part was approved on the meeting. | 07:57 |
| dtantsur | that being said, pick whichever you like | 07:57 |
| dtantsur | mmm, sorry, the 2nd part was approved | 07:57 |
| janders | I like both :) | 07:57 |
| janders | thank you dtantsur | 07:57 |
| dtantsur | do both then ;) | 07:58 |
| dtantsur | the 2nd one has two tasks, feel free to only pick the one that was approved (per Ruby's comment) | 07:58 |
| *** k_mouza has quit IRC | 07:59 | |
| *** k_mouza has joined #openstack-ironic | 08:03 | |
| janders | ok! thank you dtantsur | 08:06 |
| *** alexmcleod has joined #openstack-ironic | 08:06 | |
| *** lucasagomes has joined #openstack-ironic | 08:15 | |
| dtantsur | iurygregory: now that we're running two tests per job, I wonder if we need to raise the general timeout: https://zuul.opendev.org/t/openstack/build/7af5f601f6b84c15a5c796ee031c7104/log/job-output.txt#37946 | 08:20 |
| iurygregory | damm tempest | 08:20 |
| iurygregory | why you do this! | 08:21 |
| iurygregory | ?!* | 08:21 |
| *** janders has quit IRC | 08:21 | |
| * iurygregory facepalm | 08:22 | |
| iurygregory | the timeout of the ironic-lib jobs are 7200 | 08:22 |
| dtantsur | iurygregory: it's not tempest, it's zuul timeout | 08:22 |
| iurygregory | dtantsur, yeah I thought it was tempest | 08:22 |
| iurygregory | =) | 08:22 |
| dtantsur | I'd honestly crank it all the way up and let the more granular timeouts decide | 08:23 |
| iurygregory | yeah I will remove the timeout from the jobs and it will get from ironic-base timeout: 10800 | 08:23 |
| iurygregory | wdyt? | 08:23 |
| *** yaawang has quit IRC | 08:24 | |
| *** janders has joined #openstack-ironic | 08:24 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic-lib master: Update ironic-lib jobs https://review.opendev.org/741105 | 08:25 |
| iurygregory | ok this should take care of the timeout | 08:25 |
| dtantsur | iurygregory: I think it's a good idea to leave only one large timeout | 08:25 |
| iurygregory | dtantsur, yup and we have it on ironic-base | 08:25 |
| iurygregory | and we can't even override if we want =D | 08:26 |
| iurygregory | only for smaller values (not higher than 10800) | 08:26 |
| *** rcernin has quit IRC | 08:27 | |
| *** yaawang has joined #openstack-ironic | 08:31 | |
| *** ociuhandu_ has quit IRC | 08:33 | |
| *** ociuhandu has joined #openstack-ironic | 08:33 | |
| *** derekh has joined #openstack-ironic | 08:36 | |
| *** Lucas_Gray has joined #openstack-ironic | 08:45 | |
| *** mkrai has joined #openstack-ironic | 08:45 | |
| *** rcernin has joined #openstack-ironic | 08:48 | |
| dtantsur | iurygregory: I've realized there could be misunderstanding: the job I linked initially was not from ironic-lib | 08:52 |
| dtantsur | it's the pxe_ipmitool job from ironic | 08:52 |
| iurygregory | facepalm =( | 08:52 |
| dtantsur | it has timeout of 5400, which is very low | 08:53 |
| iurygregory | I didn't look at the name of the job =( | 08:53 |
| iurygregory | I will push a patch for it | 08:53 |
| dtantsur | thanks! | 08:53 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Wipe agent token during reboot or power off https://review.opendev.org/739964 | 08:54 |
| iurygregory | dtantsur, any problem if I use the default for all jobs that have 5400? | 08:55 |
| iurygregory | I would say we should only use the default from the ironic-base since is the maximum | 08:56 |
| iurygregory | so we won't endup with random failures due to job timeout (only tempest in some weird cases) | 08:56 |
| *** aedc has joined #openstack-ironic | 08:58 | |
| dtantsur | iurygregory: I'd honestly use the default everywhere :) but at least for these jobs | 08:59 |
| iurygregory | dtantsur, yup =) doing this | 08:59 |
| openstackgerrit | Merged openstack/ironic master: Add knob for read-only and "erase_devices" https://review.opendev.org/740731 | 09:02 |
| *** mkrai has quit IRC | 09:03 | |
| *** mkrai has joined #openstack-ironic | 09:03 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: Use default timeout for all jobs https://review.opendev.org/741123 | 09:03 |
| iurygregory | dtantsur, boom ^ | 09:04 |
| dtantsur | thx! | 09:05 |
| *** rcernin has quit IRC | 09:17 | |
| janders | is there an easy way to install reno on CentOS8? Unlike FC32, I don't seem to be able to dnf install it. Should I resort to pip? | 09:33 |
| *** ricolin_ has quit IRC | 09:49 | |
| rpittau | janders: AFAIK the rpm is available for Fedora, not Centos, probably a user install with pip is the best way | 09:51 |
| janders | thanks rpittau | 09:52 |
| janders | I jumped the gun and pip-installed system wide. Is that likely to cause issues? | 09:52 |
| rpittau | btw the package version is quite outdated even in F32 | 09:52 |
| rpittau | I would never do a system installation with pip, it may cause issues with unwanted dependencies | 09:53 |
| rpittau | it can easily break an operating system | 09:53 |
| *** mkrai has quit IRC | 09:58 | |
| *** uzumaki has joined #openstack-ironic | 10:00 | |
| uzumaki | good afternoon ironic! o/ | 10:00 |
| uzumaki | I got disconnected the past few hours, has anybody responded to my text about triaging? I might've missed that | 10:01 |
| janders | hi uzumaki o/ | 10:01 |
| janders | rpittau: noted, thank you | 10:01 |
| rpittau | uzumaki: thanks for the triaging, what was the question about? I can't find it in the backscroll | 10:03 |
| dtantsur | janders: `pip install --user reno` | 10:04 |
| *** mkrai has joined #openstack-ironic | 10:05 | |
| janders | dtantsur thank you. Just uninstalled the system-wide bits and reinstalled with --user | 10:05 |
| uzumaki | where do I put the triaged stories, like, the documentation says into the "RFE Review" section of the whiteboard, couldn'd find the section | 10:05 |
| uzumaki | also I saw a couple stories that can be marked closed | 10:05 |
| rpittau | uzumaki: to mark a story as 'closed' you just need to put all its tasks to 'merged' or 'invalid' | 10:06 |
| uzumaki | how do I do that? | 10:07 |
| uzumaki | I mean, I also wanted to bring it to somebody's quick notice before I close them off | 10:07 |
| openstackgerrit | Merged openstack/ironic-python-agent master: Ignore devices with size 0 when collecting inventory https://review.opendev.org/740309 | 10:07 |
| rpittau | each story has different tasks, they can be in status like 'to do', 'review' | 10:07 |
| *** k_mouza has quit IRC | 10:07 | |
| uzumaki | I guess what I mean is, can anybody change the status on those tasks? | 10:08 |
| rpittau | oh in that case I usually don't add the triage tag but the needs-more-info one, and add a comment askign if that can be closed | 10:08 |
| rpittau | uzumaki: yeah, anyone | 10:08 |
| uzumaki | oh I see. Well, I saw this very interesting case of a story which had a 'temporary proposal' change created on Gerrit, apparently for some discussion, which was then later ignored/discarded, and because of that the whole story is left as 'review' whereas all the action itmes have beem merged. | 10:09 |
| rpittau | uzumaki: there's not "RFE Review" in the whiteboard, that is in the weekly meeting page | 10:09 |
| rpittau | uzumaki: link ? | 10:10 |
| *** k_mouza has joined #openstack-ironic | 10:10 | |
| uzumaki | rpittau, https://storyboard.openstack.org/#!/story/2006178 | 10:10 |
| rpittau | uzumaki: that story is definitely a good candidate for being closed, and the relative still open patch to be abandoned, considering that there's no activity since 1 year | 10:12 |
| uzumaki | yes, I have 2 other candidates for closure, unlike this one, there has been no progress on them for a long time as well | 10:13 |
| uzumaki | https://storyboard.openstack.org/#!/story/2006828 | 10:13 |
| uzumaki | https://storyboard.openstack.org/#!/story/2006253 | 10:14 |
| uzumaki | https://storyboard.openstack.org/#!/story/2006047 | 10:14 |
| uzumaki | I guess there's 3 | 10:14 |
| rpittau | the hardest point is to understand if the issues are still valid or ot | 10:15 |
| rpittau | you can always ask in the story or try to do an analysis yourself | 10:15 |
| uzumaki | I agree. That's the only thing that needs to be considered before closing them off | 10:15 |
| rpittau | if the related patches are stuck since a while, it's also a viable way | 10:16 |
| rpittau | ask if the people assigned are still working on it | 10:16 |
| rpittau | then we can review that in a couple of weeks/a month | 10:16 |
| rpittau | if you ask for more info, do not add the triaged tag | 10:16 |
| uzumaki | got it, and the ones with no work at all? sitting stale for ~a year? | 10:17 |
| rpittau | that is where we need to understand if the bug/rfe is still valid, maybe it was fixed in a different story | 10:17 |
| rpittau | if it's not entirely clear, asking for info it's ok as well | 10:18 |
| rpittau | and then closing later | 10:18 |
| uzumaki | yes, I'll remove the triaged label and ask for more info | 10:18 |
| rpittau | cool, thanks :) | 10:18 |
| uzumaki | \o/ | 10:18 |
| *** yaawang has quit IRC | 10:24 | |
| *** sshnaidm is now known as sshnaidm|afk | 10:24 | |
| *** yaawang has joined #openstack-ironic | 10:25 | |
| janders | wrapping up for the day/night. Have a good one o/ | 11:01 |
| iurygregory | ironic-lib is fixed https://review.opendev.org/#/c/741105/ \o/ | 11:13 |
| patchbot | patch 741105 - ironic-lib - Update ironic-lib jobs - 2 patch sets | 11:13 |
| *** uzumaki has quit IRC | 11:29 | |
| *** rcernin has joined #openstack-ironic | 11:31 | |
| *** Lucas_Gray has quit IRC | 11:32 | |
| *** Lucas_Gray has joined #openstack-ironic | 11:46 | |
| *** rcernin has quit IRC | 11:47 | |
| *** Lucas_Gray has quit IRC | 11:58 | |
| *** Lucas_Gray has joined #openstack-ironic | 11:59 | |
| dtantsur | rpittau: https://review.opendev.org/#/c/741123/ and https://review.opendev.org/#/c/741105/ please | 12:00 |
| patchbot | patch 741123 - ironic - Use default timeout for all jobs - 1 patch set | 12:00 |
| patchbot | patch 741105 - ironic-lib - Update ironic-lib jobs - 2 patch sets | 12:00 |
| *** mkrai has quit IRC | 12:03 | |
| *** dsneddon has joined #openstack-ironic | 12:03 | |
| *** Lucas_Gray has quit IRC | 12:04 | |
| *** dsneddon has quit IRC | 12:11 | |
| *** tkajinam has quit IRC | 12:24 | |
| *** Lucas_Gray has joined #openstack-ironic | 12:30 | |
| *** rloo has joined #openstack-ironic | 12:35 | |
| *** vapjes has joined #openstack-ironic | 12:35 | |
| rpittau | dtantsur: done! | 12:40 |
| openstackgerrit | Merged openstack/ironic master: Use min_command_interval when ironic does IPMI retries https://review.opendev.org/740466 | 12:43 |
| dtantsur | thanks! oh, the CI behaves again, neat! | 12:45 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 12:50 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Do not validate driver on changing non-driver fields https://review.opendev.org/739699 | 12:52 |
| *** sshnaidm|afk is now known as sshnaidm | 12:53 | |
| dtantsur | here goes the 13th recheck | 12:53 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic stable/ussuri: Use min_command_interval when ironic does IPMI retries https://review.opendev.org/741169 | 12:54 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent stable/ussuri: Ignore devices with size 0 when collecting inventory https://review.opendev.org/741171 | 12:54 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent stable/train: Ignore devices with size 0 when collecting inventory https://review.opendev.org/741172 | 12:55 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost stable/ussuri: Removing libselinux-python package from Fedora dependencies https://review.opendev.org/741173 | 12:55 |
| openstackgerrit | Bob Fournier proposed openstack/ironic stable/ussuri: Use min_command_interval when ironic does IPMI retries https://review.opendev.org/741169 | 12:56 |
| dtantsur | great minds think alike :D | 12:57 |
| bfournie | dtantsur: yep :-) I missed that you did that already | 12:58 |
| dtantsur | with 2 minutes of difference, yeah :) I haven't touched train since it has a conflict | 12:58 |
| bfournie | dtantsur: so the original patch wasn't backported to Train, I'm trying to remember if that was intentional or just hadn't been done | 12:59 |
| dtantsur | mmm, I see. yeah, not sure either, it probably should have backported.. | 13:00 |
| bfournie | dtantsur: I will do it, since the default will stay the same to use ipmitool retries there will be no functional affect | 13:01 |
| openstackgerrit | Merged openstack/bifrost master: Do not clone repos with source_install==false https://review.opendev.org/739930 | 13:16 |
| TheJulia | brraaaaiinnnsss | 13:18 |
| TheJulia | is the gate any happier? | 13:18 |
| rpittau | never happy, just slightly less unhappy | 13:19 |
| TheJulia | well I'll make some coffee and revise the content type fix for ironic-inspector, that should make things at least 10% better | 13:20 |
| openstackgerrit | Riccardo Pittau proposed openstack/sushy master: Set min version of tox to 3.2.1 https://review.opendev.org/741184 | 13:22 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Set min version of tox to 3.2.1 https://review.opendev.org/741185 | 13:25 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: Set min version of tox to 3.2.1 https://review.opendev.org/741186 | 13:26 |
| *** jj123 has joined #openstack-ironic | 13:27 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent master: Set min version of tox to 3.2.1 https://review.opendev.org/741187 | 13:28 |
| dtantsur | morning TheJulia | 13:28 |
| openstackgerrit | Riccardo Pittau proposed openstack/sushy-tools master: Set min version of tox to 3.2.1 https://review.opendev.org/741188 | 13:30 |
| openstackgerrit | Riccardo Pittau proposed openstack/sushy-cli master: Set min version of tox to 3.2.1 https://review.opendev.org/741189 | 13:31 |
| openstackgerrit | Riccardo Pittau proposed openstack/python-ironicclient master: Set min version of tox to 3.2.1 https://review.opendev.org/741190 | 13:31 |
| openstackgerrit | Riccardo Pittau proposed openstack/python-ironic-inspector-client master: Set min version of tox to 3.2.1 https://review.opendev.org/741191 | 13:32 |
| iurygregory | spam time | 13:33 |
| iurygregory | good morning TheJulia =) | 13:33 |
| openstackgerrit | Riccardo Pittau proposed openstack/virtualbmc master: Set min version of tox to 3.2.1 https://review.opendev.org/741192 | 13:34 |
| jj123 | I'm starting on prototyping a Redfish-compliant BMC for use with Ironic. My server is not a typical BMC target and I prefer to use a Linux distribution rather than Yocto for development flexibility, so OpenBMC doesn't look helpful. I'm planning to use the sushy-tools dynamic emulator as a starting point, but appreciate any other suggestions. | 13:35 |
| openstackgerrit | Riccardo Pittau proposed openstack/bifrost master: Set min version of tox to 3.2.1 https://review.opendev.org/741193 | 13:36 |
| openstackgerrit | Riccardo Pittau proposed openstack/tenks master: Set min version of tox to 3.2.1 https://review.opendev.org/741194 | 13:37 |
| *** stendulker has joined #openstack-ironic | 13:37 | |
| openstackgerrit | Riccardo Pittau proposed openstack/networking-baremetal master: Set min version of tox to 3.2.1 https://review.opendev.org/741195 | 13:38 |
| TheJulia | jj123: That is awesome to hear, unfortunately we don't have any advice in that area, but if we can help in any way with sushy tools, please don't hesitate to ask | 13:38 |
| TheJulia | and good morning everyone | 13:38 |
| jj123 | TheJulia thanks :) | 13:38 |
| openstackgerrit | Riccardo Pittau proposed openstack/networking-generic-switch master: Set min version of tox to 3.2.1 https://review.opendev.org/741196 | 13:39 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-prometheus-exporter master: Set min version of tox to 3.2.1 https://review.opendev.org/741200 | 13:42 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-tempest-plugin master: Set min version of tox to 3.2.1 https://review.opendev.org/741201 | 13:43 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent-builder master: Set min version of tox to 3.2.1 https://review.opendev.org/741202 | 13:44 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Allow deleting nodes with a broken driver https://review.opendev.org/739702 | 13:45 |
| openstackgerrit | Riccardo Pittau proposed openstack/metalsmith master: Set min version of tox to 3.2.1 https://review.opendev.org/741203 | 13:45 |
| stendulker | Good morning TheJulia | 13:45 |
| stendulker | TheJulia: One quick question, what is the rationale for inband deploy step supporting only steps from 'deploy' interface? | 13:46 |
| TheJulia | stendulker: I don't think there was one beyond mental connotation to connect the two | 13:48 |
| TheJulia | with the word deploy | 13:49 |
| TheJulia | so a logical extension is raid | 13:49 |
| stendulker | so is management for say firmware_update | 13:49 |
| stendulker | Deploy steps do support all kinds of OOB steps | 13:49 |
| TheJulia | management as well, but operators that ask for firmware updates as part of deployment will have to wait an exceptionally long time so they will need to have an environment set to support htis | 13:50 |
| *** tzumainn has joined #openstack-ironic | 13:50 | |
| stendulker | that's true | 13:52 |
| stendulker | i think its its broken for raid as well | 13:52 |
| stendulker | The get_deploy_steps() implementation in agent_base.py looks into agent cached steps. | 13:53 |
| stendulker | That takes care of 'agent' interface. All other interfaces falls to base.py implementation which does not look for agent cached steps | 13:54 |
| TheJulia | dtantsur: mgoddard: any thoughts on this ^ | 13:55 |
| stendulker | Hence the doubt, if we want to implement raid related inband steps as part of 'deploy' interface | 13:55 |
| TheJulia | well, I'm a little worried that your concerns are coming up now considering mgoddard started down the path to support raid on deploy | 13:55 |
| TheJulia | when you say cached steps, do you mean the hard coded "we know about thse" or the in node cache from the beginning of the step operations? | 13:57 |
| stendulker | I mean the steps retrieved from the agent | 13:58 |
| stendulker | and storaed as agent_cached_deploy_steps in driver_internal_info | 13:59 |
| TheJulia | wheee one of flask's documented ways of forcing the content type just doesnt work | 13:59 |
| *** rajinir has joined #openstack-ironic | 14:00 | |
| iurygregory | maybe the version of flask? | 14:01 |
| TheJulia | dunno, tried the way listed in someone's blog post from like 2018 and it worked like a champ | 14:01 |
| TheJulia | just a little uglier | 14:02 |
| *** Goneri has joined #openstack-ironic | 14:11 | |
| dtantsur | stendulker, TheJulia, it may be a simple issue to fix | 14:13 |
| openstackgerrit | Merged openstack/bifrost stable/ussuri: Removing libselinux-python package from Fedora dependencies https://review.opendev.org/741173 | 14:15 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: [WIP] Refactor 'when' statements https://review.opendev.org/741217 | 14:16 |
| dtantsur | rpittau: this is a huge rabbit hole ^^ | 14:16 |
| dtantsur | early input welcome | 14:16 |
| rpittau | yeah | 14:16 |
| rpittau | :) | 14:16 |
| * rpittau tea time | 14:17 | |
| TheJulia | I recommed Acme portable holes instead of rabbit holes | 14:17 |
| TheJulia | Now the question is, how many people get the above cartoon reference | 14:17 |
| rpittau | json-rpc hole? | 14:18 |
| dtantsur | I had to look it up :) | 14:18 |
| rpittau | poor Wile :) | 14:18 |
| dtantsur | stendulker, TheJulia, okay, yeah, the deploy interface only fetches 'deploy' interface steps. sort of logical.. AgentRAID will need an update. | 14:21 |
| TheJulia | I guess the logical next question is what about the dedicated raid interfaces | 14:23 |
| *** abdysn has quit IRC | 14:26 | |
| dtantsur | each vendor should decide, what to do about their RAID interfaces (I'm fixing AgentRAID now) | 14:31 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Implement get_deploy_steps for AgentRAID https://review.opendev.org/741225 | 14:33 |
| dtantsur | like this ^^^ | 14:33 |
| openstackgerrit | Jakub Libosvar proposed openstack/ironic master: Explicitly set jobs to ML2/OVS https://review.opendev.org/739945 | 14:33 |
| openstackgerrit | Jakub Libosvar proposed openstack/ironic master: Explicitly set jobs to ML2/OVS https://review.opendev.org/739945 | 14:34 |
| openstackgerrit | Jakub Libosvar proposed openstack/ironic master: DNM: Just test the patch under with devstack change https://review.opendev.org/741226 | 14:35 |
| iurygregory | wow | 14:35 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-tempest-plugin master: POC deploy-time RAID https://review.opendev.org/741227 | 14:39 |
| TheJulia | wheeeee flask hides things quite well | 14:41 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-inspector master: Respond so a ContentType is included https://review.opendev.org/740344 | 14:45 |
| TheJulia | dtantsur: as requested ^ | 14:45 |
| dtantsur | thanks! | 14:46 |
| TheJulia | That plus the cpu core patches should fix up inspectors CI issues | 14:46 |
| TheJulia | which will increase ironic's pass rate as well | 14:46 |
| TheJulia | at least, hopefully | 14:46 |
| TheJulia | Taking the dog out because he is asking... for the third time this morning :( | 14:46 |
| TheJulia | Anyone have any words about the impending baremetal whitepaper, there is an ask for quotes from contributors | 14:49 |
| dtantsur | have we intentionally omitted support for passing deploy steps with 'baremetal node deploy'? | 14:50 |
| rpittau | TheJulia: quotes? :) | 14:50 |
| TheJulia | dtantsur: intentionally unlikely | 14:50 |
| dtantsur | mgoddard: maybe you remember ^^? | 14:50 |
| TheJulia | dtantsur: so many different areas that the entire set of changes touched that it may have just gone missing along the way | 14:51 |
| dtantsur | oh so true | 14:51 |
| dtantsur | but let's check with Mark, maybe it was actually intentional? | 14:51 |
| TheJulia | rpittau: quotes like "I believe xyz" or "ironic will rule the baremetal world" or whatever! | 14:51 |
| rpittau | ok, got it | 14:52 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-inspector master: Respond so a ContentType is included https://review.opendev.org/740344 | 14:59 |
| TheJulia | dtantsur: ^^ changed to helper method | 14:59 |
| *** diurnalist has joined #openstack-ironic | 15:02 | |
| dtantsur | TheJulia: I rather imagined something like `return _make_empty_response()` | 15:06 |
| TheJulia | yeah, that makes way more sense | 15:07 |
| TheJulia | except | 15:07 |
| TheJulia | two different return codes to support | 15:07 |
| TheJulia | one no-content and one accepted | 15:07 |
| stendulker | Thank you dtantsur for the clarification and a patch :) | 15:08 |
| dtantsur | np | 15:08 |
| dtantsur | TheJulia: mm, yeah, so _make_empty_response(204) | 15:09 |
| TheJulia | yup, I'll revise again shortly | 15:09 |
| *** mkrai has joined #openstack-ironic | 15:09 | |
| TheJulia | lucasagomes: Hey, spotted your reply about the newer neutron plugin going away eventually | 15:12 |
| lucasagomes | TheJulia, hi there. Yeah, it was decided in the last PTG because, the new module still lacks features with the legacy one and nobody is working on it | 15:13 |
| TheJulia | lucasagomes: which is frankly a disaster for ironic. Do you know if there would be any receptiveness for there to be any cross-gating against ironic so neutron's in-devstack plugin doesn't kill our gate? | 15:13 |
| TheJulia | lucasagomes: I suspect part of the problem is nobody wants to review it | 15:13 |
| lucasagomes | TheJulia, are you using the new module ? | 15:13 |
| TheJulia | lucasagomes: patches against it in the past have sat for months even when we've had people asking for reviews | 15:14 |
| TheJulia | lucasagomes: well, we use the newer names which is supposed to invoke the newer module if memory serves | 15:14 |
| lucasagomes | TheJulia, right, I can double-check because I believe that if there's any q-* service enable the module will then defaults to legacy | 15:15 |
| TheJulia | lucasagomes: that is a huge behavior change then from what the default was | 15:15 |
| lucasagomes | TheJulia, but that said, there is a ironic job on the devstack gate so, we will make sure we do not break it | 15:15 |
| TheJulia | but if it has been done then \o/ | 15:15 |
| TheJulia | and I won't go drinking heavily | 15:15 |
| lucasagomes | lol | 15:15 |
| lucasagomes | to be honest, I don't know if there's anyone working on deleting that module and deprecating the "legacy" one at the moment | 15:16 |
| * TheJulia goes and looks at that job since I don't remember an ironic job on devstack | 15:16 | |
| lucasagomes | TheJulia, ironic-tempest-ipa-wholedisk-bios-agent_ipmitool-tinyipa | 15:16 |
| lucasagomes | it's non-voting tho, this job should be fairly stable no ? Maybe we can make it voting | 15:17 |
| TheJulia | yeah, its been stable for ages | 15:17 |
| TheJulia | it does need to be renamed | 15:17 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Decompose the core deploy step on iscsi and ansible deploy https://review.opendev.org/731180 | 15:17 |
| TheJulia | I can propose a change and see if they will be kind | 15:18 |
| lucasagomes | TheJulia++ | 15:18 |
| lucasagomes | yeah let's make it voting | 15:18 |
| TheJulia | does anyone know why we never renamed the ironic-tempest-ipa-wholedisk-bios-agent_ipmitool-tinyipa job? | 15:20 |
| TheJulia | looks like it is everyone's favorite cross-gating job | 15:22 |
| *** dsneddon has joined #openstack-ironic | 15:25 | |
| dtantsur | that's why | 15:28 |
| TheJulia | yeah | 15:33 |
| TheJulia | well, might as well alias | 15:33 |
| TheJulia | protip: don't try to make zuul.conf changes... on a very out of date copy of master | 15:33 |
| TheJulia | *boom* | 15:33 |
| * TheJulia looks at what git tried to do to the yaml and wonders if git like... went and smoked some illegal substances and before returning "git stash pop" | 15:34 | |
| TheJulia | err, ironic-jobs.yaml | 15:35 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Remove old driver name from cross-gating job https://review.opendev.org/741254 | 15:39 |
| *** mraineri has joined #openstack-ironic | 15:40 | |
| rpittau | TheJulia: that job is the last one still using tinyipa, that's why | 15:41 |
| TheJulia | We shouldn't eradicate it then | 15:41 |
| rpittau | I'm trying here https://review.opendev.org/734759 | 15:41 |
| patchbot | patch 734759 - ironic - Convert ironic-tempest-ipa-wholedisk-bios-agent_ip... - 11 patch sets | 15:41 |
| rpittau | networkmanager disagrees | 15:41 |
| TheJulia | why would networkmanager care? | 15:42 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic master: DO NOT REVIEW: Test Ironic with DevStack's neutron-legacy plugin https://review.opendev.org/741256 | 15:42 |
| rpittau | because it uses the second (unconfigured) network card to lookup the node, and fails miserably | 15:42 |
| TheJulia | rpittau: and that is the basis for dropping tinyipa support? | 15:43 |
| rpittau | that is the last tinyipa job left in ironic | 15:43 |
| TheJulia | I still don't understand, because if tinycore 11 uses network manager and for some reason only the apparently unregistered with ironic nic is getting sent over, that is not a good reason to drop tinyipa, that seems like a fundimental set of bugs in both our we setup the test environment and ipa as to what macs are sent and when | 15:45 |
| TheJulia | maybe I'm just really confused? | 15:45 |
| rpittau | TheJulia: the patch that I linked uses centos8 image, with NetworkManager | 15:46 |
| TheJulia | so, the network manager issue is with dib images. | 15:46 |
| TheJulia | which means it is a completely disjointed issue from what I'm seeking to understand | 15:46 |
| rpittau | ok, I was answering to the question "does anyone know why we never renamed the ironic-tempest-ipa-wholedisk-bios-agent_ipmitool-tinyipa job?" | 15:47 |
| TheJulia | I was more wanting to eradicate "agent_ipmitool" | 15:47 |
| *** k_mouza has quit IRC | 15:55 | |
| TheJulia | lucasagomes: I am likely way too verbose on the commit message, but https://review.opendev.org/#/c/741265/ | 16:02 |
| patchbot | patch 741265 - devstack - Make ironic cross gating voting - 1 patch set | 16:02 |
| lucasagomes | TheJulia, thanks! Will take a look | 16:03 |
| *** mkrai has quit IRC | 16:07 | |
| *** k_mouza has joined #openstack-ironic | 16:11 | |
| *** lucasagomes has quit IRC | 16:15 | |
| *** lbotell57 has joined #openstack-ironic | 16:15 | |
| *** k_mouza has quit IRC | 16:16 | |
| iurygregory | TheJulia, -1 ironic-tempest-wholedisk-bios-ipmi-direct-tinyipa doesn't exist =) | 16:18 |
| TheJulia | zuuuuul!!! | 16:19 |
| TheJulia | I guess it is only in its local repo's understanding of the running config? | 16:19 |
| iurygregory | at least I got nothing from http://codesearch.openstack.org/?q=ironic-tempest-wholedisk-bios-ipmi-direct-tinyipa&i=nope&files=&repos= | 16:20 |
| TheJulia | iurygregory: I posted a chance to put it in place to ironic | 16:21 |
| TheJulia | renaming the old name | 16:21 |
| iurygregory | oh =) | 16:21 |
| openstackgerrit | vinay kumar muddu proposed openstack/ironic master: Decouple the ISO creation logic from redfish https://review.opendev.org/739174 | 16:21 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-inspector master: Respond so a ContentType is included https://review.opendev.org/740344 | 16:22 |
| TheJulia | dtantsur: ^^ | 16:22 |
| *** k_mouza has joined #openstack-ironic | 16:24 | |
| *** k_mouza has quit IRC | 16:24 | |
| *** k_mouza has joined #openstack-ironic | 16:24 | |
| TheJulia | If any other core reviewers want to take a quick look at https://review.opendev.org/#/c/740344 it should help make the inspector jobs much happier which will mean we're only fighting the race condition in ramdisk decompression. Patches already approved for that but they won't likely be able to merge without 740334 merged | 16:25 |
| patchbot | patch 740344 - ironic-inspector - Respond so a ContentType is included - 6 patch sets | 16:25 |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic master: Fixes to skip validation of in-band deploy steps before agent boot https://review.opendev.org/741267 | 16:25 |
| iurygregory | TheJulia, I can take a look after I finish my exercises =) | 16:28 |
| openstackgerrit | Bob Fournier proposed openstack/ironic stable/train: New configuration parameter to use ipmitool retries https://review.opendev.org/741268 | 16:29 |
| *** k_mouza has quit IRC | 16:29 | |
| *** ociuhandu_ has joined #openstack-ironic | 16:36 | |
| *** ociuhandu has quit IRC | 16:39 | |
| *** ociuhandu_ has quit IRC | 16:40 | |
| *** stendulker has quit IRC | 16:43 | |
| TheJulia | Has anyone heard from shukdev recently? | 16:45 |
| * dtantsur hasn't for a good while | 16:46 | |
| TheJulia | I know he jumped into irc like 3 months ago and then nothing | 16:48 |
| *** dking has quit IRC | 16:51 | |
| TheJulia | I found his email address and sent him a quick message | 16:59 |
| *** derekh has quit IRC | 17:00 | |
| openstackgerrit | Verification of a change to openstack/ironic-lib failed: Update ironic-lib jobs https://review.opendev.org/741105 | 17:01 |
| *** lbotell57 has quit IRC | 17:05 | |
| arne_wiebalck | bye everyone o/ | 17:11 |
| TheJulia | goodnight | 17:14 |
| *** sri_ has joined #openstack-ironic | 17:15 | |
| *** dtantsur is now known as dtantsur|afk | 17:18 | |
| dtantsur|afk | o/ | 17:18 |
| *** gyee has joined #openstack-ironic | 17:30 | |
| TheJulia | looks like most jobs are now failing | 17:33 |
| *** Lucas_Gray has quit IRC | 17:41 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Use default timeout for all jobs https://review.opendev.org/741123 | 18:05 |
| *** larsks has joined #openstack-ironic | 18:14 | |
| *** ociuhandu has joined #openstack-ironic | 18:46 | |
| *** ociuhandu has quit IRC | 18:50 | |
| TheJulia | lets hold off on rechecks of anything unless we know it is going to help fix the gate | 18:52 |
| *** rmcall has quit IRC | 19:22 | |
| *** rmcall has joined #openstack-ironic | 19:23 | |
| *** rmcallis has joined #openstack-ironic | 19:25 | |
| *** rmcall has quit IRC | 19:28 | |
| *** rmcallis__ has joined #openstack-ironic | 19:28 | |
| *** rmcallis has quit IRC | 19:31 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Add `get_node_network_data` to Neutron NetworkInterface https://review.opendev.org/688896 | 19:44 |
| *** dking has joined #openstack-ironic | 19:57 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Add json and param parsing to args https://review.opendev.org/704489 | 19:57 |
| dking | I'm installing a whole-disk-image with local_boot using Bifrost, but the root partition is only the size of the image. How would I make the partition grow to fill the full disk? Is that something I would build into the image itself, something that I configure within Bifrost, or should I not be using a whole-disk-image? | 20:04 |
| clarkb | dking: you can use a tool like growroot which will install a init script/unit that will frow filesystems on boot | 20:11 |
| clarkb | if you are using disk image builder there is a growroot element which should configure it properly for your image | 20:12 |
| dking | clarkb: Thank you. That may be what I need. It looks like it grows it on first boot? I'm using Packer with qemu to modify the image after DIB but before deploy. Do you know if that cause it to grow for the Packer run? | 20:17 |
| clarkb | if packer starts the operating system and runs the init scripts then yes that will likely grow it during the packer run | 20:18 |
| dking | Okay. I might give it a shot and see. I suppose that if that doesn't work, then I could just install growpart in Packer or figure out what DIB does with it and redo that there. | 20:23 |
| TheJulia | *sigh* CI hates us | 20:30 |
| TheJulia | bfournie: I thought https://review.opendev.org/#/c/741268/1 was too low?!? | 20:36 |
| patchbot | patch 741268 - ironic (stable/train) - New configuration parameter to use ipmitool retries - 1 patch set | 20:36 |
| TheJulia | iurygregory: when we spoke earlier, you mentioned something about multiple VMs... was that a patch? | 20:41 |
| bfournie | TheJulia: that one has to merge to Train first, then I can merge the follow-on, But this has no affect on default as it still using ipmitool retries by default | 20:44 |
| TheJulia | stevebaker: http://paste.openstack.org/show/795959/ | 20:45 |
| TheJulia | bfournie: ack | 20:45 |
| bfournie | TheJulia: in other words, that patch followed by backport of https://review.opendev.org/#/c/741268/. | 20:45 |
| patchbot | patch 741268 - ironic (stable/train) - New configuration parameter to use ipmitool retries - 1 patch set | 20:45 |
| TheJulia | bfournie: is the next one already proposed? | 20:45 |
| TheJulia | Whiteboard updated with latest CI nightmare | 20:47 |
| stevebaker | TheJulia: thanks | 20:49 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Use min_command_interval when ironic does IPMI retries https://review.opendev.org/741169 | 20:54 |
| TheJulia | it _looks_ like.... we don't think it is a partition image | 21:01 |
| TheJulia | we think it is a wholedisk image | 21:01 |
| *** gyee has quit IRC | 21:10 | |
| *** gyee has joined #openstack-ironic | 21:11 | |
| TheJulia | I'm starting to suspect devstack broke us | 21:11 |
| TheJulia | https://review.opendev.org/#/c/735966/9/functions | 21:11 |
| patchbot | patch 735966 - devstack - Use glance import workflow for creating image (MERGED) - 9 patch sets | 21:11 |
| TheJulia | yup, they blew up partition image support | 21:22 |
| *** rmcall has joined #openstack-ironic | 21:27 | |
| *** rmcallis__ has quit IRC | 21:28 | |
| *** rmcall has quit IRC | 21:30 | |
| *** rmcall has joined #openstack-ironic | 21:31 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: DNM: Test of devstack fix https://review.opendev.org/741323 | 21:32 |
| *** rmcall has quit IRC | 21:35 | |
| *** rmcallis has joined #openstack-ironic | 21:36 | |
| *** rmcallis has quit IRC | 21:42 | |
| TheJulia | Okay, we've got competing fixes going in for devstack :) | 21:44 |
| TheJulia | or at least being reviewed | 21:44 |
| *** jj123 has quit IRC | 21:52 | |
| *** rloo has quit IRC | 22:12 | |
| *** fdegir has quit IRC | 22:16 | |
| *** fdegir has joined #openstack-ironic | 22:17 | |
| *** logan- has quit IRC | 22:17 | |
| *** rcernin has joined #openstack-ironic | 22:19 | |
| *** logan- has joined #openstack-ironic | 22:19 | |
| *** rcernin has quit IRC | 22:20 | |
| *** rcernin has joined #openstack-ironic | 22:21 | |
| *** vapjes has quit IRC | 22:33 | |
| *** hjensas has quit IRC | 22:44 | |
| janders | g'day ironic o/ | 22:45 |
| openstackgerrit | Merged openstack/ironic stable/ussuri: Stop using md5 for __repr__ of objects https://review.opendev.org/740329 | 22:48 |
| openstackgerrit | Merged openstack/ironic stable/train: Stop using md5 for __repr__ of objects https://review.opendev.org/740330 | 22:49 |
| iurygregory | TheJulia, sorry I took a nap and I just woke up =) I've mentioned that we need to increase the number of VM's for the jobs since are running wholedisk / partition for ironic-lib it worked in the begin https://review.opendev.org/#/c/741105/ I haven't dig into the failures to see what is happening | 22:50 |
| patchbot | patch 741105 - ironic-lib - Update ironic-lib jobs - 2 patch sets | 22:50 |
| * iurygregory will read the openstack-qa in the morning (can't process info atm hehe) | 22:51 | |
| iurygregory | we also decided to use the default timeout for all jobs (since some had 5400 as limit) https://review.opendev.org/741123 | 22:52 |
| patchbot | patch 741123 - ironic - Use default timeout for all jobs - 1 patch set | 22:52 |
| *** tkajinam has joined #openstack-ironic | 22:55 | |
| *** tkajinam has quit IRC | 22:55 | |
| *** tkajinam has joined #openstack-ironic | 22:58 | |
| dking | clarkb: Thank you very much for the help. I've tested a few times, and it seems that simply adding growroot to the DIB build works fine in deployment, even though I am using Packer with qemu to boot and provision the deploy image. | 23:14 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!