| *** janders has joined #openstack-ironic | 00:01 | |
| *** hoonetorg has quit IRC | 00:10 | |
| *** janders has quit IRC | 00:18 | |
| *** rcernin has quit IRC | 00:42 | |
| *** rcernin has joined #openstack-ironic | 00:46 | |
| *** janders has joined #openstack-ironic | 01:36 | |
| *** ricolin has joined #openstack-ironic | 01:40 | |
| *** tzumainn has quit IRC | 01:51 | |
| *** rcernin has quit IRC | 02:13 | |
| *** antonym has quit IRC | 02:15 | |
| *** rcernin has joined #openstack-ironic | 02:15 | |
| *** antonym has joined #openstack-ironic | 02:31 | |
| *** rcernin has quit IRC | 02:33 | |
| *** rcernin has joined #openstack-ironic | 02:55 | |
| *** mkrai has joined #openstack-ironic | 03:02 | |
| *** ociuhandu has joined #openstack-ironic | 03:16 | |
| *** ociuhandu has quit IRC | 03:20 | |
| *** janders has quit IRC | 03:27 | |
| *** mkrai has quit IRC | 03:28 | |
| *** mkrai has joined #openstack-ironic | 03:29 | |
| *** hjensas has quit IRC | 03:42 | |
| *** hjensas has joined #openstack-ironic | 03:43 | |
| *** diurnalist has quit IRC | 04:19 | |
| dking | I don't know if anybody is currently around, but if so, would anybody know what would be the API states/provision targets I can use that would be the equivalent to "openstack baremetal node manage" and "openstack baremetal node provide"? I'm hoping that I can bring a node to available using simple curl commands. I can do the deploy with '{ "target": "active"}' if I can get it to available. | 04:57 |
|---|---|---|
| *** abdysn has joined #openstack-ironic | 05:05 | |
| *** diurnalist has joined #openstack-ironic | 05:32 | |
| *** mkrai has quit IRC | 05:49 | |
| *** mkrai has joined #openstack-ironic | 05:59 | |
| *** diurnalist has quit IRC | 06:04 | |
| *** ianw is now known as ianw_pto | 06:38 | |
| *** trident has quit IRC | 06:39 | |
| *** trident has joined #openstack-ironic | 06:42 | |
| *** janders has joined #openstack-ironic | 06:48 | |
| *** rpittau has joined #openstack-ironic | 06:50 | |
| rpittau | good morning ironic! o/ | 06:50 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Add documentation on how to propose a release https://review.opendev.org/739678 | 06:57 |
| janders | g'day rpittau o/ | 06:58 |
| rpittau | hey janders :) | 06:58 |
| *** rcernin has quit IRC | 06:59 | |
| *** rcernin has joined #openstack-ironic | 07:06 | |
| *** mkrai has quit IRC | 07:14 | |
| iurygregory | good morning rpittau janders and Ironic o/ | 07:21 |
| iurygregory | janders, FYI we have open a bz for the mariadb problem https://bugzilla.redhat.com/show_bug.cgi?id=1854637 | 07:23 |
| openstack | bugzilla.redhat.com bug 1854637 in mariadb "bifrost fails when running ironic-dbsync create_schema" [High,New] - Assigned to mschorm | 07:23 |
| *** rpittau has quit IRC | 07:23 | |
| janders | thank you iurygregory | 07:23 |
| janders | I suppose you ran into the same issue in your testing? | 07:24 |
| iurygregory | yeah | 07:24 |
| janders | interesting! I suppose I now need to give my old trusty laptop a good screen clean with apologies | 07:24 |
| iurygregory | hehehe | 07:25 |
| janders | I was suspecting it's too old/slow and hence causing this | 07:25 |
| *** rpittau has joined #openstack-ironic | 07:25 | |
| iurygregory | if you want to add any info to the bug feel free | 07:26 |
| *** rpittau has quit IRC | 07:31 | |
| *** rcernin has quit IRC | 07:41 | |
| *** rpittau has joined #openstack-ironic | 07:42 | |
| *** mkrai has joined #openstack-ironic | 07:47 | |
| *** hjensas has quit IRC | 07:49 | |
| *** dougsz has joined #openstack-ironic | 07:51 | |
| *** dtantsur|afk is now known as dtantsur | 08:10 | |
| *** ociuhandu has joined #openstack-ironic | 08:10 | |
| dtantsur | morning ironic | 08:10 |
| *** Lucas_Gray has joined #openstack-ironic | 08:11 | |
| rpittau | hey dtantsur :) | 08:11 |
| dtantsur | janders: congratulations, you've found a serious bug on the 2nd day of your employment :) | 08:11 |
| janders | haha :) thank you dtantsur | 08:11 |
| *** mbuil has quit IRC | 08:13 | |
| rpittau | janders, are you going to propose the patch for libselinux in bifrost? | 08:15 |
| *** Wryhder has joined #openstack-ironic | 08:16 | |
| *** Lucas_Gray has quit IRC | 08:16 | |
| *** mkrai has quit IRC | 08:17 | |
| *** Wryhder is now known as Lucas_Gray | 08:17 | |
| *** mkrai has joined #openstack-ironic | 08:19 | |
| *** lucasagomes has joined #openstack-ironic | 08:19 | |
| *** derekh has joined #openstack-ironic | 08:19 | |
| *** tkajinam has quit IRC | 08:23 | |
| dtantsur | rpittau: could you check https://review.opendev.org/#/c/728315/ please? I think it's a good candidate for a last-minute change in the release | 08:23 |
| patchbot | patch 728315 - ironic - Provide a path to set explicit ipxe bootloaders - 9 patch sets | 08:23 |
| dtantsur | meanwhile, we can release ironic-inspector. please don't forget about bugfix/* branches (I commented on bifrost already). | 08:24 |
| rpittau | dtantsur, yep, I thought 8.2 was ok based on https://review.opendev.org/739089, for example we removed support for old Fedora and Debian | 08:29 |
| patchbot | patch 739089 - bifrost - Prepare release notes for release 8.2 (MERGED) - 1 patch set | 08:29 |
| *** k_mouza has joined #openstack-ironic | 08:30 | |
| dtantsur | rpittau: 8.2 is probably fine, yes, although I do remember Julia mentioning a major release | 08:31 |
| dtantsur | rpittau: also, which patch do you have in mind in https://review.opendev.org/#/c/739671/ ? | 08:32 |
| patchbot | patch 739671 - releases - Release ironic 13.0.5 for train - 1 patch set | 08:32 |
| dtantsur | I don't see anything ready to merge for train | 08:32 |
| rpittau | arf wrong patch | 08:32 |
| *** diurnalist has joined #openstack-ironic | 09:00 | |
| *** hjensas has joined #openstack-ironic | 09:01 | |
| rpittau | dtantsur, we probably need to add the stable-branch-type std-with-versions to the bifrost yaml file | 09:02 |
| *** mbuil has joined #openstack-ironic | 09:03 | |
| dtantsur | rpittau: likely so | 09:03 |
| rpittau | ok, I'll push an update | 09:03 |
| *** diurnalist has quit IRC | 09:04 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Do not clone repos with source_install==false https://review.opendev.org/739930 | 09:05 |
| openstackgerrit | Merged openstack/ironic master: Log when a node should be fast-track-able but it's not https://review.opendev.org/739189 | 09:06 |
| openstackgerrit | Merged openstack/ironic master: Do not verify error messages from jsonpatch in unit tests https://review.opendev.org/739803 | 09:06 |
| dtantsur | prometheanfire: ^^ | 09:08 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: agent_base: support inserting in-band deploy steps https://review.opendev.org/737695 | 09:11 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Decompose the core deploy step of the direct deploy https://review.opendev.org/698309 | 09:11 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Decompose the core deploy step on iscsi and ansible deploy https://review.opendev.org/731180 | 09:11 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Agent power interface https://review.opendev.org/735591 | 09:15 |
| dtantsur | okay, rebase party is over :) | 09:15 |
| ftarasenko | arne_wiebalck: Hi Arne! Thank you for your software raid implementation, it works well. Still testing, but results are good. Have one more question for you - why do we try to find rootfs partition by UUID and PARTUUID, but do not try to find it by LABEL? It seems much easier and 100% predictible just to set LABEL for new image rootfs. | 09:21 |
| *** rcernin has joined #openstack-ironic | 09:26 | |
| *** rcernin has quit IRC | 09:41 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Do not clone repos with source_install==false https://review.opendev.org/739930 | 09:54 |
| *** derekh has quit IRC | 09:57 | |
| *** derekh has joined #openstack-ironic | 09:58 | |
| janders | rpittau: yes, I am intending to propose the libselinux "patch" for FC32. | 10:12 |
| janders | IRC newbie question: is there a "nice" way to respond to a message from an hour ago (think clicking reply in Signal/Whatsapp/Slack) | 10:12 |
| rpittau | janders, ok, thanks :) | 10:14 |
| openstackgerrit | Merged openstack/ironic master: Add documentation on how to propose a release https://review.opendev.org/739678 | 10:14 |
| rpittau | janders, usually what you just did is ok :) | 10:14 |
| janders | ACK :) | 10:14 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 10:19 |
| janders | I'm about to set myself up development server. In the past I successfuly used all-in-one OpenStacks for this use case - can grab a VM of any image/flavour with any comms in seconds. Would it make sense to do this in ironic/metal3 context, or is this a bad idea? | 10:20 |
| janders | I suppose one of the key bits is whether vBMC can work with OpenStack or is it a pure-libvirt beast? | 10:20 |
| openstackgerrit | Jakub Libosvar proposed openstack/ironic master: Explicitly set jobs to ML2/OVS https://review.opendev.org/739945 | 10:24 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Increase vm ram size in uefi netbooted jobs https://review.opendev.org/739728 | 10:25 |
| *** janders has quit IRC | 10:28 | |
| *** janders has joined #openstack-ironic | 10:28 | |
| *** mkrai has quit IRC | 10:42 | |
| *** arne_wiebalck has joined #openstack-ironic | 11:04 | |
| arne_wiebalck | Good morning, ironic! | 11:05 |
| * arne_wiebalck was hoping irccloud would come back at some point | 11:05 | |
| dtantsur | arne_wiebalck: ftarasenko had a question for you this morning | 11:12 |
| *** dsneddon has quit IRC | 11:25 | |
| *** dsneddon has joined #openstack-ironic | 11:26 | |
| arne_wiebalck | dtantsur: thanks! | 11:30 |
| openstackgerrit | Jakub Libosvar proposed openstack/ironic master: Explicitly set jobs to ML2/OVS https://review.opendev.org/739945 | 11:31 |
| arne_wiebalck | ftarasenko: I have no backlog atm ... can I still help? (I will be afk for an hour or so now, though) | 11:31 |
| maelk | Hi! I have a question related to IPA token generation / re-generation. Metal3 is broken right now due to an issue related to the token. Our nodes go through the introspection without issue. They are then powered off. Once we start the provisioning, ironic powers them on, and right away we can see in the logs : `An agent token generation request is | 11:34 |
| maelk | being refused as one is already present for node xxx` and after that the heartbeats are failing because IPA is not presenting a token. So I am trying to understand the process wrt to the token cleaning. When is the token supposed to be cleared in this case ? | 11:34 |
| dtantsur | maelk: hey, is it related to that problem of unnecessary powering nodes on after inspection? | 11:36 |
| maelk | well it seems that it started failing after we powered them off | 11:36 |
| maelk | but with saw the same problem with fast track | 11:36 |
| maelk | you said at some point that even with fast track, the nodes were rebooting once, right dtantsur? | 11:38 |
| dtantsur | maelk: it has been recently fixed, but probably hasn't hit the repos yet | 11:39 |
| dtantsur | I'll get a tea real quick and will think more about it | 11:39 |
| dtantsur | do you have any logs? | 11:39 |
| maelk | sure, plenty, all our CI runs are failing since yesterday!! :D | 11:40 |
| dtantsur | could you link me to one of them please? :) | 11:40 |
| dtantsur | and the change that caused the failure if you're aware of one | 11:40 |
| maelk | https://jenkins.nordix.org/view/Airship/job/airship_master_v1a3_integration_test_ubuntu/242/artifact/logs-jenkins-airship_master_v1a3_integration_test_ubuntu-242.tgz in there, go to docker and there are ironic and inspector logs | 11:41 |
| dtantsur | maelk: do we collect ramdisk logs? | 11:41 |
| * dtantsur -> tea, brb | 11:41 | |
| maelk | I think what caused the failure is : https://github.com/metal3-io/baremetal-operator/commit/9e02a1e0b70da0965b75b2e701dc00001619214f | 11:41 |
| maelk | yes, the ramdisk logs are under the qemu folders | 11:42 |
| rpittau | mmm the logs under the qemu folder are empty | 11:46 |
| maelk | dtantsur sorry it looks like the console is empty. that's the only thing we get from the vms :/ so we don't have the ramdisk logs | 11:46 |
| rpittau | we need to check the version of the ipa in the ramdisk | 11:46 |
| rpittau | and I think what we see in the inspector logs was recently fixed | 11:46 |
| maelk | we get our ramdisk from https://images.rdoproject.org/centos8/master/rdo_trunk/current-tripleo/ | 11:47 |
| rpittau | this -> AttributeError: 'Node' object has no attribute 'uuid' | 11:48 |
| maelk | yes, that's fixed | 11:48 |
| rpittau | yeah, I'm checking that | 11:48 |
| maelk | but that is not the actual error that throws us off | 11:48 |
| maelk | our problem is with the IPA tokens | 11:48 |
| rpittau | yep :) | 11:48 |
| maelk | I could try to run the ramdisk with console=ttyS0 to get the console and see if we get anything from IPA | 11:52 |
| rpittau | that would be great | 11:52 |
| dtantsur | what I'm worried is this situation: we do inspection, then do a lookup (receive a token), then start heartbeating. Then we can rebooted for deployment and the repeated lookup does not receive a token. Boom! | 11:54 |
| dtantsur | rpittau, maelk, how likely is ^^? | 11:54 |
| rpittau | that looks like what's happening | 11:54 |
| dtantsur | maelk: what exactly does 'online: false' do? | 11:55 |
| maelk | it tells BMO to ensure the node is turned off | 11:55 |
| maelk | afaik, we should not have any intermediary reboot | 11:55 |
| dtantsur | maelk: if BMO kills the ramdisk after a lookup, that's what you'll get | 11:56 |
| dtantsur | I guess we could wipe the agent token on reboot.. | 11:56 |
| maelk | it does not really kill it, it set the power state in ironic | 11:57 |
| dtantsur | well, same effect, I'm afraid | 11:57 |
| maelk | isn't ironic turning off the node after introspection anyways ? | 11:57 |
| maelk | because we do notuse fast track | 11:57 |
| maelk | I though the problem we were previously hitting was because ironic was powering off the node after introspection and then BMO requesting it to be powered on again | 11:58 |
| dtantsur | maelk: we are telling ironic not to | 12:00 |
| dtantsur | and we should really use fast track | 12:00 |
| dtantsur | maelk: do you have a bug report for this issue? I'd like to link it to my code change. | 12:01 |
| maelk | not yet, because I don't really understand what causes it | 12:02 |
| maelk | I don't get at what time the extra reboot would be happening between the introspection and the deployment | 12:03 |
| dtantsur | maelk: you said BMO will ensure it's turned off, no? | 12:04 |
| maelk | yes. does ironic otherwise turn off a node after introspection ? | 12:04 |
| maelk | afaiu the node is powered on, introspected, then powered off, until it is powered on again for deployment (cleaning etc.) Did I get something wrong ? | 12:05 |
| dtantsur | maelk: ironic won't power off the node (that's how we configure it in metal3 for fast track), so BMO ends up powering it off instead. Causing this. | 12:07 |
| maelk | but fast track has been disabled for months in Metal3 | 12:07 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Wipe agent token during reboot or power off https://review.opendev.org/739964 | 12:07 |
| dtantsur | maelk: it's not fully disabled IIRC | 12:08 |
| dtantsur | anyway, is there a chance you could try this ^^^ patch? | 12:08 |
| *** rcernin has joined #openstack-ironic | 12:08 | |
| openstackgerrit | Merged openstack/ironic master: Correct Redfish boot once fallback conditional https://review.opendev.org/737892 | 12:08 |
| openstackgerrit | Merged openstack/ironic master: Provide a path to set explicit ipxe bootloaders https://review.opendev.org/728315 | 12:08 |
| maelk | dtantsur I'll try to test it | 12:09 |
| maelk | thank you | 12:10 |
| maelk | If we prevent BMO from turning off the nodes, that should fix it, no ? | 12:11 |
| maelk | as a temporary work around | 12:11 |
| dtantsur | maelk: try this as a workaround: https://github.com/metal3-io/ironic-image/pull/173 | 12:13 |
| maelk | there is still something I cannot get. I'm sorry for that. Considering the case where BMO does not turn the nodes off after introspection, how did we end up with a reboot right after introspection ? (causing IPA to issue a lookup query and then going through introspection) | 12:13 |
| maelk | oh, your PR explained that :D | 12:14 |
| dtantsur | maelk: not sure I get the question. The sequence as I see it is this: 1. inspection finishes, power_off is set to false, so the node proceeds to lookup, 2. BMO shuts down the node, 3. Deployment starts, but the token is already used. | 12:14 |
| dtantsur | good :) | 12:14 |
| maelk | actually, no I misunderstood | 12:15 |
| *** rcernin has quit IRC | 12:16 | |
| maelk | ah so IPA does a lookup anyways after introspection ? | 12:16 |
| dtantsur | maelk: yup. we configure it to use fast-track unconditionally. mmm, I may expand my fix further. | 12:16 |
| maelk | so fast track was really baked in and when we disabled it in the ironic config, it resulted in a half way broken deployment | 12:17 |
| dtantsur | maelk: pretty much (although it's because of another potential issue in ironic). see the updated PR. | 12:19 |
| maelk | What is the impact of not giving the api url ? | 12:20 |
| dtantsur | maelk: the presence of this URL enables fast-track as far as the agent is concerned | 12:21 |
| maelk | oh, got it | 12:21 |
| maelk | thanks | 12:21 |
| dtantsur | maelk: let's see what the CI says about the PR. if it gets green, we're on the right track :) | 12:22 |
| maelk | ok, I triggered it. so now the node will be powered off after introspection, and when powered on at deployment time, it will be able to do the lookup and get a token. that's the expected result, right ? | 12:24 |
| maelk | shouldn't we remove the parameter from inspector.ipxe too ? | 12:24 |
| dtantsur | maelk: yep | 12:24 |
| dtantsur | and yes, we should, but it's less pressing | 12:25 |
| *** hoonetorg has joined #openstack-ironic | 12:25 | |
| dtantsur | (the code path in inspector.ipxe is only hit on discovery in the metal3 context) | 12:25 |
| dtantsur | I'll update the PR once we get the CI results | 12:25 |
| dtantsur | rpittau: did you have a chance to check my comment on https://review.opendev.org/#/c/739714/ ? | 12:25 |
| patchbot | patch 739714 - releases - Release bifrost 8.2.0 for victoria - 3 patch sets | 12:25 |
| dtantsur | maelk: I'd appreciate if you could file a github issue with at least our high-level findings | 12:28 |
| maelk | ok, I'll file the issue. | 12:28 |
| rpittau | dtantsur, I didn't see that, looking at it now | 12:28 |
| rpittau | and.. weird | 12:28 |
| maelk | One more question dtantsur about your last comment. how come we hit it only on discovery ? when we boot over PXE, we'll directly land there, no ? | 12:29 |
| dtantsur | maelk: we're getting a bit deep into technical details.. but there are two ways inspection is started in ironic: | 12:29 |
| dtantsur | "managed" inspection is started by ironic itself and ironic-inspector only processes the data. it does NOT use inspector.ipxe, but rather files generated by ironic. this is the default for metal3. | 12:30 |
| dtantsur | the "unmanaged" code path works when new nodes are discovered (or when a node does not have ports). we shouldn't hit it in metal3 by default. | 12:31 |
| dtantsur | the latter uses inspector.ipxe | 12:31 |
| maelk | ok. I was trying to understand how, when using the managed inspection, the node boots over PXE and loads the ramdisk with files from Ironic. dnsmasq does not seem to be serving anything generated by ironic, or is it just because the tftp directory is shared ? | 12:45 |
| maelk | that might be a question too specific to Metal3, we can move the discussion to slack there if you wish | 12:45 |
| dtantsur | maelk: it's not metal3 specific, we use managed inspection in bifrost as well. DHCP is served the same way, but the iPXE files are generated by ironic rather than using the static inspector.ipxe file. | 12:46 |
| maelk | ok, got it! thanks a lot for the detailed explanations | 12:49 |
| dtantsur | np! when you get a bug link, I can mark the PR as closing it (I need to update a few things anyway) | 12:50 |
| *** yaawang has quit IRC | 12:54 | |
| *** rpittau has quit IRC | 12:57 | |
| *** yaawang has joined #openstack-ironic | 12:58 | |
| openstackgerrit | Merged openstack/ironic-python-agent master: Refactor part of image module https://review.opendev.org/739251 | 13:00 |
| openstackgerrit | Merged openstack/ironic master: Follow up of enforce autospec in test_manager https://review.opendev.org/739063 | 13:00 |
| *** rpittau has joined #openstack-ironic | 13:00 | |
| openstackgerrit | Merged openstack/ironic master: Enforce autospec in test_task_manager https://review.opendev.org/739758 | 13:00 |
| openstackgerrit | Merged openstack/ironic-inspector master: Add release version to release notes https://review.opendev.org/739814 | 13:00 |
| *** rloo has joined #openstack-ironic | 13:06 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Remove bifrost-openstack-ci-prep role https://review.opendev.org/739978 | 13:07 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Do not clone repos with source_install==false https://review.opendev.org/739930 | 13:07 |
| dtantsur | maelk: it passes \o/ | 13:08 |
| *** dougsz has quit IRC | 13:17 | |
| openstackgerrit | Merged openstack/sushy-tools master: Use victoria tests https://review.opendev.org/739173 | 13:26 |
| *** dougsz has joined #openstack-ironic | 13:31 | |
| derekh | ERROR ironic.conductor.manager [req-749f7c91-ac3d-4453-aaef-0bb7e3207cbf - - - - -] Suspicious activity detected for node 9515b037-4c0e-4188-818f-7c32342fd8ce when attempting to heartbeat. Heartbeat request has been rejected as the version of ironic-python-agent indicated in the heartbeat operation should support agent token functionality. | 13:34 |
| derekh | ^^ do I need to turn on agent token on IPA ? | 13:34 |
| iurygregory | the best error we have in agent token XD | 13:35 |
| dtantsur | derekh: where are you seeing this? | 13:39 |
| derekh | dtantsur: I've tried to run metal3-dev-env, which I think is using master | 13:41 |
| derekh | openstack-ironic-conductor-15.1.0-0.20200629133120.e461e36.el8.noarch and IPA from https://images.rdoproject.org/centos8/master/rdo_trunk/current-tripleo/ | 13:42 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Guard conductor from consuming all of the ram https://review.opendev.org/726483 | 13:42 |
| dtantsur | derekh: it may be the issue fixed by https://github.com/metal3-io/ironic-image/pull/173 | 13:46 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent-builder master: [WIP] Use tinycore 11 to build tinyipa https://review.opendev.org/711647 | 13:47 |
| *** tzumainn has joined #openstack-ironic | 13:48 | |
| *** Goneri has joined #openstack-ironic | 13:48 | |
| derekh | dtantsur: ack, I'll see if I can figure out how to run that PR in this env | 13:48 |
| *** mkrai has joined #openstack-ironic | 13:50 | |
| rpittau | love when can't reproduce CI errors locally.... | 13:52 |
| iurygregory | i live the name of the branch "bugfix/8.2" | 13:55 |
| iurygregory | like* | 13:55 |
| *** rpittau_ has joined #openstack-ironic | 13:56 | |
| dtantsur | okay, we've got a release and a branch, yay! | 13:56 |
| *** pas-ha has joined #openstack-ironic | 13:56 | |
| rpittau_ | \o/ | 13:56 |
| dtantsur | if anybody likes removing code: https://review.opendev.org/#/c/739978/ | 13:57 |
| patchbot | patch 739978 - bifrost - Remove bifrost-openstack-ci-prep role - 1 patch set | 13:57 |
| *** rpittau has quit IRC | 13:58 | |
| dtantsur | rpittau_: I +1'ed your release requests. I think ironic is ready to go as well | 14:00 |
| rpittau_ | dtantsur, great! :) | 14:01 |
| iurygregory | dtantsur, rpittau_ do you have any ideas why ironic-tempest-ipa-partition-pxe_ipmitool doesn't like the wholedisk basic_ops test? .-. https://review.opendev.org/714453 I'm trying to reproduce locally | 14:11 |
| patchbot | patch 714453 - ironic - add tempest boot_mode config - 17 patch sets | 14:11 |
| iurygregory | should we add a new condition to skip the wholedisk basic_ops for this job only? | 14:11 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: ISO ramdisk virtual media test enablement https://review.opendev.org/736383 | 14:12 |
| dtantsur | iurygregory: I'm intending to take a look later today. Maybe insufficient configuration in devstack? | 14:22 |
| iurygregory | dtantsur, yeah, I'm trying to debug but no luck yet | 14:23 |
| maelk | dtantsur your PR in metal3 was already merged before I had time to write the issue. I'll still write it, as a reference. Thank you for the fix, looks like the CI will go back to green! | 14:24 |
| dtantsur | maelk: yep, no worries. do we need to cherry-pick it to openshift? /me never remembers the process | 14:24 |
| dtantsur | rpittau_, iurygregory, joining the compat matrix discussion? https://meetpad.opendev.org/ironic | 14:31 |
| *** abdysn has quit IRC | 14:31 | |
| iurygregory | dtantsur, 1min | 14:31 |
| rpittau_ | I should be there already | 14:31 |
| rpittau_ | oh... it crashed | 14:32 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: add tempest boot_mode config https://review.opendev.org/714453 | 14:39 |
| iurygregory | let's see if it helps, the only difference from config I could spot from ironic-tempest-ipa-partition-pxe_ipmitool vs ironic-tempest-ipa-partition-uefi-pxe_ipmitool | 14:40 |
| dtantsur | iurygregory: yeah, but we need cleaning enabled somewhere.. | 14:46 |
| dtantsur | otherwise we're not testing it at all | 14:46 |
| iurygregory | dtantsur, well it wasn't testing.. https://0d061d36abfdb86463d9-e8b640c415c002ebe2f9d8685ad47e75.ssl.cf2.rackcdn.com/714453/17/check/ironic-tempest-ipa-partition-pxe_ipmitool/34e6fc6/testr_results.html | 14:47 |
| iurygregory | O.o | 14:48 |
| dtantsur | iurygregory: sorry, I'm not catching why you think it wasn't | 14:48 |
| iurygregory | the config didn't lead to any tempest test for cleaning... | 14:48 |
| *** priteau has joined #openstack-ironic | 14:48 | |
| iurygregory | IRONIC_AUTOMATED_CLEAN_ENABLED we don't have it set to True anywhere O.o | 14:49 |
| dtantsur | it's on by default in devstack maybe? | 14:50 |
| dtantsur | it's not a separate test, it happens before tempest | 14:50 |
| iurygregory | yeah it's true by default | 14:50 |
| iurygregory | oh ok, I though we had a specific test in tempest | 14:50 |
| dtantsur | iurygregory: lemme suggest something | 14:50 |
| iurygregory | dtantsur, sure =) | 14:51 |
| dtantsur | we're switching from running 1 test per job to running 2? | 14:51 |
| dtantsur | that means, there is cleaning between them? and maybe the 2nd test does not find a free node? | 14:51 |
| dtantsur | so, maybe we need to change the job to having 2 test nodes? | 14:51 |
| iurygregory | aha | 14:52 |
| iurygregory | thats a good point | 14:52 |
| iurygregory | oh maybe we should move cleaning to a job that have more VM's... | 14:53 |
| iurygregory | not sure if 2 VM's of 3GB would be ok, wdyt? | 14:54 |
| openstackgerrit | Merged openstack/ironic stable/ussuri: Increase vm ram size in uefi netbooted jobs https://review.opendev.org/739728 | 14:55 |
| dtantsur | iurygregory: we can change this job to tinyIPA if it proves a problem | 14:55 |
| iurygregory | dtantsur, ack, let me update for the IRONIC_VM_COUNT to 2 | 14:56 |
| dtantsur | iurygregory: and let's maybe set IRONIC_AUTOMATED_CLEAN_ENABLED to True explicitly to make it clear what is happening? | 14:57 |
| dtantsur | and leave a comment to this extent? | 14:57 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: add tempest boot_mode config https://review.opendev.org/714453 | 14:57 |
| iurygregory | dtantsur, will do | 14:57 |
| iurygregory | let me see how it goes with 2 VMs, since we may need to put tinyipa... | 14:58 |
| iurygregory | tks for the help dtantsur o/ | 14:58 |
| *** k_mouza has quit IRC | 15:00 | |
| * iurygregory had to drop the landlord came to show the flat to some people | 15:01 | |
| *** k_mouza has joined #openstack-ironic | 15:02 | |
| *** diurnalist has joined #openstack-ironic | 15:02 | |
| *** priteau has quit IRC | 15:04 | |
| derekh | dtantsur: ya, that PR looks like it fixed it, thanks | 15:16 |
| dtantsur | awesome | 15:16 |
| * derekh wasn't fast enough to add a review | 15:17 | |
| openstackgerrit | Merged openstack/bifrost master: Remove bifrost-openstack-ci-prep role https://review.opendev.org/739978 | 15:20 |
| openstackgerrit | Merged openstack/ironic-tempest-plugin master: Improve test_baremetal_basic_ops https://review.opendev.org/735960 | 15:32 |
| *** k_mouza has quit IRC | 15:36 | |
| arne_wiebalck | We recently came across 2 cases where we had nodes which had multiple BMC channels with valid IP addresses. The problem is that the IPA takes the first one when trying to find the node in Ironic. Would it make sense to try all? | 15:40 |
| arne_wiebalck | The "solution" in our case was to disable the other channels. | 15:40 |
| *** k_mouza has joined #openstack-ironic | 15:41 | |
| dtantsur | arne_wiebalck: it probably does make sense to try all, although I cannot imagine why anybody would need many channels.. | 15:42 |
| arne_wiebalck | dtantsur: we do not need them either, it is just they have one assigned ... it is a misconfiguration, I was just wondering if we should fail (as we do now) or try our best to not fail (which would mean to try all IPs) | 15:46 |
| dtantsur | ideally, the latter. but it requires quite a few changes on both IPA and inspector sides | 15:51 |
| *** gyee has joined #openstack-ironic | 15:52 | |
| arne_wiebalck | dtantsur: yeah ... I guess the BMC address in hardware_info would become a list, we would need a loop when calling back, ... | 15:55 |
| *** ociuhandu_ has joined #openstack-ironic | 15:55 | |
| dtantsur | yeah, and keep backward compatibility while changing it :) do you think a lot of people would hit that? | 15:55 |
| arne_wiebalck | dtantsur: how would this break backwards compatibility? | 15:56 |
| arne_wiebalck | dtantsur: wouldn't the IPA just call back twice? | 15:56 |
| arne_wiebalck | dtantsur: anyway, we don't have to do the work now :) | 15:57 |
| dtantsur | I mean, if you change bmc_address to a list, old inspectors will object | 15:57 |
| *** ociuhandu has quit IRC | 15:58 | |
| arne_wiebalck | dtantsur: oh ... can't the IPA call back for each, so the list stays local? | 15:58 |
| arne_wiebalck | dtantsur: I have not looked at the callback code tbh, just at the code which gets the IP from the BMC | 15:58 |
| *** gyee has quit IRC | 15:59 | |
| dtantsur | it may have implications.. dunno | 15:59 |
| arne_wiebalck | dtantsur: yeah, not sure how much of an issue this is | 15:59 |
| arne_wiebalck | dtantsur: anyway, thanks for your thoughts! | 15:59 |
| *** ociuhandu_ has quit IRC | 15:59 | |
| *** dtantsur is now known as dtantsur|afk | 16:01 | |
| dtantsur|afk | o/ | 16:01 |
| *** ociuhandu has joined #openstack-ironic | 16:12 | |
| *** ociuhandu has quit IRC | 16:18 | |
| *** lucasagomes has quit IRC | 16:18 | |
| *** dougsz has quit IRC | 16:32 | |
| *** mkrai has quit IRC | 16:38 | |
| *** derekh has quit IRC | 16:41 | |
| *** akahat has quit IRC | 16:46 | |
| *** ociuhandu has joined #openstack-ironic | 16:47 | |
| *** ociuhandu has quit IRC | 16:53 | |
| *** k_mouza has quit IRC | 16:53 | |
| *** Lucas_Gray has quit IRC | 16:57 | |
| *** gyee has joined #openstack-ironic | 17:07 | |
| *** pas-ha has quit IRC | 17:07 | |
| *** ociuhandu has joined #openstack-ironic | 17:16 | |
| *** rh-jelabarre has quit IRC | 17:17 | |
| *** ociuhandu has quit IRC | 17:23 | |
| iurygregory | dtantsur|afk, it worked by increasing the VM COUNT \o/ | 17:25 |
| iurygregory | I will add the IRONIC_AUTOMATED_CLEAN_ENABLED and update the description of the job =) | 17:26 |
| *** akahat has joined #openstack-ironic | 17:27 | |
| *** ricolin has quit IRC | 18:00 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Stop using md5 for __repr__ of objects https://review.opendev.org/739735 | 18:21 |
| *** sshnaidm|ruck is now known as sshnaidm|afk | 18:25 | |
| *** ociuhandu has joined #openstack-ironic | 18:35 | |
| *** ociuhandu has quit IRC | 18:41 | |
| *** rpittau_ has quit IRC | 19:01 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: add tempest boot_mode config https://review.opendev.org/714453 | 19:07 |
| *** rh-jelabarre has joined #openstack-ironic | 19:15 | |
| *** gmann_ has joined #openstack-ironic | 19:17 | |
| *** gmann_ is now known as gmann | 19:18 | |
| *** arne_wiebalck_ has joined #openstack-ironic | 19:30 | |
| *** donnyd has joined #openstack-ironic | 19:31 | |
| *** mnaser has joined #openstack-ironic | 19:32 | |
| *** portdirect has joined #openstack-ironic | 19:32 | |
| *** TheJulia has joined #openstack-ironic | 19:35 | |
| *** jungleboyj_ has joined #openstack-ironic | 19:35 | |
| *** buhman has joined #openstack-ironic | 19:36 | |
| *** v1k0d3n has joined #openstack-ironic | 19:36 | |
| *** ab-a has quit IRC | 19:37 | |
| *** ab-a has joined #openstack-ironic | 19:38 | |
| *** rpittau has joined #openstack-ironic | 19:40 | |
| *** arne_wiebalck has quit IRC | 19:41 | |
| *** arne_wiebalck_ is now known as arne_wiebalck | 19:41 | |
| *** tonyb[m] has joined #openstack-ironic | 19:44 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Minor agent token log handling https://review.opendev.org/736214 | 19:49 |
| *** rpioso has joined #openstack-ironic | 19:54 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: add tempest boot_mode config https://review.opendev.org/714453 | 19:56 |
| *** ildikov has joined #openstack-ironic | 19:56 | |
| iurygregory | this should fix our CI since ironic-tempest-ipa-wholedisk-bios-agent_ipmitool-tinyipa ironic-tempest-ipa-partition-pxe_ipmitool are failing | 19:56 |
| iurygregory | at least increasing the number of VMs for ironic-tempest-ipa-partition-pxe_ipmitool helped so I've increased to 2 in ironic-tempest-ipa-wholedisk-bios-agent_ipmitool-tinyipa | 19:57 |
| *** johnhaan has joined #openstack-ironic | 20:06 | |
| *** m1p has joined #openstack-ironic | 20:09 | |
| *** johnsom_ has joined #openstack-ironic | 20:10 | |
| *** pas-ha has joined #openstack-ironic | 20:10 | |
| *** phrobb has joined #openstack-ironic | 20:14 | |
| *** samueldmq has joined #openstack-ironic | 20:16 | |
| *** erbarr has joined #openstack-ironic | 20:16 | |
| *** seongsoocho has joined #openstack-ironic | 20:17 | |
| *** vkmc_ has joined #openstack-ironic | 20:17 | |
| *** andrein has joined #openstack-ironic | 20:21 | |
| *** NobodyCam has joined #openstack-ironic | 20:36 | |
| *** vdrok has joined #openstack-ironic | 20:44 | |
| *** Lucas_Gray has joined #openstack-ironic | 20:52 | |
| *** mnaser is now known as mnaser|ic | 20:52 | |
| TheJulia | \o/ irccloud is back | 20:55 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Do not validate driver on changing non-driver fields https://review.opendev.org/739699 | 20:58 |
| *** mnaser|ic has quit IRC | 21:08 | |
| *** mnaser|ic has joined #openstack-ironic | 21:08 | |
| *** mnaser|ic has quit IRC | 21:08 | |
| *** mnaser|ic has joined #openstack-ironic | 21:08 | |
| *** mnaser|ic is now known as vexxhost | 21:08 | |
| *** vexxhost is now known as mnaser | 21:14 | |
| *** mnaser is now known as mnaser|ic | 21:14 | |
| *** johnsom_ is now known as johnsom | 21:19 | |
| *** johnsom has joined #openstack-ironic | 21:19 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Add a driver compatability matrix https://review.opendev.org/733716 | 21:26 |
| openstackgerrit | Verification of a change to openstack/python-ironicclient failed: Allow to pass global request id for remaining objects https://review.opendev.org/725941 | 21:27 |
| openstackgerrit | Verification of a change to openstack/python-ironicclient failed: Add release note regarding global_request_id https://review.opendev.org/732590 | 21:27 |
| *** aedc has quit IRC | 21:33 | |
| TheJulia | iurygregory: I feel like you mentioned you were working on updating https://review.opendev.org/#/c/688896 ? is that right? | 21:39 |
| patchbot | patch 688896 - ironic - Add `get_node_network_data` to Neutron NetworkInte... - 23 patch sets | 21:39 |
| iurygregory | TheJulia, right, I've updated the things Dmitry asked =) | 21:39 |
| iurygregory | at least I think I did all the necessary changes =) | 21:39 |
| TheJulia | oh heh, I'm looking at rev 22, not 23 | 21:40 |
| iurygregory | ops XD | 21:41 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Software RAID: don't try to set local_gb to MAX https://review.opendev.org/739488 | 21:53 |
| *** uzumaki has joined #openstack-ironic | 21:55 | |
| *** portdirect has quit IRC | 21:56 | |
| *** portdirect has joined #openstack-ironic | 21:56 | |
| openstackgerrit | Merged openstack/ironic master: [doc] Describe how to extract an fs UUID from an image https://review.opendev.org/739027 | 21:59 |
| *** uzumaki has quit IRC | 22:03 | |
| *** sshnaidm|afk has quit IRC | 22:03 | |
| *** rloo has quit IRC | 22:14 | |
| openstackgerrit | Merged openstack/ironic-lib master: Set min version of tox to 3.2.1 https://review.opendev.org/739436 | 22:15 |
| TheJulia | hmm.. looks like the gate failure rate is doing better than it was yesterday | 22:18 |
| *** tkajinam has joined #openstack-ironic | 22:58 | |
| NobodyCam | wow fun irccloud outage ... Good afternoon all o/ | 23:00 |
| *** rcernin has joined #openstack-ironic | 23:01 | |
| *** mnaser|ic has quit IRC | 23:38 | |
| *** mnaser|ic has joined #openstack-ironic | 23:38 | |
| *** mnaser|ic has quit IRC | 23:38 | |
| *** mnaser|ic has joined #openstack-ironic | 23:38 | |
| *** mnaser|ic is now known as mnaser | 23:38 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!