| openstackgerrit | Merged openstack/ironic-inspector master: Limit inspector jobs to 1 testing VM https://review.opendev.org/753051 | 00:11 |
|---|---|---|
| openstackgerrit | Richard G. Pioso proposed x/sushy-oem-idrac master: WIP: Accommodate GET HTTP 202 status without Location https://review.opendev.org/726275 | 00:16 |
| openstackgerrit | Richard G. Pioso proposed x/sushy-oem-idrac master: Fix GET HTTP 202 status without Location https://review.opendev.org/726275 | 00:21 |
| *** jawad_axd has joined #openstack-ironic | 00:34 | |
| TheJulia | dtantsur|afk: I think it would be okay if you just +2+A https://review.opendev.org/#/c/753183 in the morning. It passed which should allow other changes to move through on ironic if they hit smaller nodes :\ | 00:38 |
| patchbot | patch 753183 - ironic - Reduce grenade node count - 1 patch set | 00:38 |
| *** jawad_axd has quit IRC | 00:38 | |
| TheJulia | dtantsur|afk: the multinode and standalone job patch following, at least as I'm fanally stepping away for the day does seem good | 00:39 |
| *** cdearborn has quit IRC | 00:41 | |
| *** Goneri has quit IRC | 00:59 | |
| *** jawad_axd has joined #openstack-ironic | 01:15 | |
| *** jawad_axd has quit IRC | 01:19 | |
| *** gyee has quit IRC | 01:22 | |
| *** jawad_axd has joined #openstack-ironic | 01:36 | |
| *** jawad_axd has quit IRC | 01:41 | |
| *** Qianbiao has joined #openstack-ironic | 01:43 | |
| *** jawad_axd has joined #openstack-ironic | 01:57 | |
| *** jawad_axd has quit IRC | 02:01 | |
| *** zzzeek has quit IRC | 02:51 | |
| *** zzzeek has joined #openstack-ironic | 02:54 | |
| *** Qianbiao has quit IRC | 03:06 | |
| *** Qianbiao has joined #openstack-ironic | 03:06 | |
| *** rh-jelabarre has quit IRC | 03:12 | |
| *** uzumaki has joined #openstack-ironic | 03:44 | |
| openstackgerrit | wu.shiming proposed openstack/python-ironicclient master: requirements: Drop os-testr https://review.opendev.org/753210 | 03:57 |
| *** uzumaki has quit IRC | 04:02 | |
| openstackgerrit | Steve Baker proposed openstack/bifrost master: Add some interfaces for development hardware types https://review.opendev.org/753213 | 04:07 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Add status_code argument to @method.expose https://review.opendev.org/743677 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Add @method.body decorator https://review.opendev.org/743678 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Duplicate trait validation with jsonschema https://review.opendev.org/745778 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: New argument validate decorator https://review.opendev.org/742795 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Utility functions for REST API JSON handling https://review.opendev.org/751160 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert allocations endpoint to plain JSON https://review.opendev.org/743679 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert bios endpoint to plain JSON https://review.opendev.org/745475 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert chassis endpoint to plain JSON https://review.opendev.org/745476 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert conductors endpoint to plain JSON https://review.opendev.org/745779 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert deploy_templates endpoint to plain JSON https://review.opendev.org/745780 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert drivers endpoint to plain JSON https://review.opendev.org/741865 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert event endpoint to plain JSON https://review.opendev.org/746623 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert nodes endpoint to plain JSON https://review.opendev.org/747820 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert ports endpoint to plain JSON https://review.opendev.org/750120 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert portgroups endpoint to plain JSON https://review.opendev.org/750864 | 04:20 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert ramdisk endpoint to plain JSON https://review.opendev.org/750857 | 04:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert volume endpoint to plain JSON https://review.opendev.org/750858 | 04:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert volume/connectors endpoint to plain JSON https://review.opendev.org/750859 | 04:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert volume/targets endpoint to plain JSON https://review.opendev.org/750860 | 04:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Expunge the internal version of WSME https://review.opendev.org/750861 | 04:21 |
| *** tzumainn has quit IRC | 04:22 | |
| TheJulia | stevebaker: bravo | 04:29 |
| stevebaker | twas a journey | 04:30 |
| *** uzumaki has joined #openstack-ironic | 04:39 | |
| *** rcernin has quit IRC | 05:14 | |
| *** Qianbiao has quit IRC | 05:20 | |
| *** jtomasek has joined #openstack-ironic | 05:24 | |
| *** rcernin has joined #openstack-ironic | 05:34 | |
| arne_wiebalck | Good morning, ironic! | 05:34 |
| *** Qianbiao has joined #openstack-ironic | 05:35 | |
| *** abdysn has joined #openstack-ironic | 05:39 | |
| *** Qianbiao has quit IRC | 05:43 | |
| *** Qianbiao has joined #openstack-ironic | 05:43 | |
| *** zzzeek has quit IRC | 05:54 | |
| *** zzzeek has joined #openstack-ironic | 05:54 | |
| *** jtomasek has quit IRC | 06:21 | |
| *** zzzeek has quit IRC | 06:26 | |
| *** zzzeek has joined #openstack-ironic | 06:28 | |
| uzumaki | bonjour arne_wiebalck ! coment ca va? | 06:39 |
| arne_wiebalck | Salut uzumaki, ca va et toi? | 06:39 |
| uzumaki | je vais bien.. merci :) | 06:40 |
| uzumaki | comment le travail? | 06:40 |
| *** Qianbiao has quit IRC | 06:51 | |
| *** Qianbiao has joined #openstack-ironic | 06:51 | |
| *** bfournie has quit IRC | 06:53 | |
| arne_wiebalck | Comme d'hab il y a toujours des petits problèmes, donc ... intéressant :) | 06:55 |
| iurygregory | good morning arne_wiebalck uzumaki and Ironic o/ | 06:55 |
| uzumaki | arne_wiebalck, ah! petits problemes interresant eh? Je vois.. | 06:56 |
| arne_wiebalck | Hey iurygregory o/ | 06:56 |
| uzumaki | comme avec moi xD aujoud'ui, je me sens c'est lundi, le fatigue, c'est la meme | 06:56 |
| uzumaki | iurygregory, morning! o/ | 06:57 |
| *** jtomasek has joined #openstack-ironic | 07:00 | |
| arne_wiebalck | uzumaki: Si tu n'as pas à résoudre les problèmes tout de suite, ils restent intéressants. Sinon, ils deviennent | 07:00 |
| arne_wiebalck | énervants bien sur :) | 07:00 |
| uzumaki | Je sais.. c'est vrai.. :D | 07:02 |
| uzumaki | quelque fois, on aime les problemes, et autre fois, on les deteste. C'est etrange! | 07:05 |
| uzumaki | J'ecoute S.O.S par Indila. C'est ma chanson preferee! <3 | 07:06 |
| *** belmoreira has joined #openstack-ironic | 07:10 | |
| *** bfournie has joined #openstack-ironic | 07:10 | |
| openstackgerrit | wu.shiming proposed openstack/bifrost master: Fix hacking min version to 3.0.1 https://review.opendev.org/753235 | 07:11 |
| *** rcernin has quit IRC | 07:13 | |
| Qianbiao | Morning Ironic. | 07:15 |
| Qianbiao | what language that is :) | 07:15 |
| Qianbiao | am I on Mars. | 07:15 |
| uzumaki | Morning Qianbiao :) that's French :D | 07:15 |
| Qianbiao | :) | 07:15 |
| uzumaki | arne_wiebalck, is from Switzerland, and I like to practice my French when we talk :D | 07:16 |
| uzumaki | It's fun! | 07:16 |
| arne_wiebalck | actually, I am from Germany, I just live in France :) | 07:16 |
| uzumaki | Oh, my bad! I do remember you work around the border area of france and switzerland? | 07:17 |
| uzumaki | tu es allemand? =O arne_wiebalck | 07:17 |
| arne_wiebalck | I work in Switzerland, yes, but live in France. And I am German, yes :) | 07:17 |
| uzumaki | Bizarreeeeee!!! :D | 07:18 |
| uzumaki | It's like one of those videos I saw, the boders between netherlands and belgium, people have a house half in one country and half in the other, with two addresses :D | 07:18 |
| uzumaki | of course, your case is stranger, ce n'est pas deux pays, c'est trois :D | 07:19 |
| Qianbiao | arne_wiebalck interesting experience. :) | 07:22 |
| arne_wiebalck | Yeah, I think there exist some houses here as well which are on the border. Like a hotel with rooms in two coutries. | 07:22 |
| *** jawad_axd has joined #openstack-ironic | 07:22 | |
| uzumaki | arne_wiebalck, be like, oh gotta to the the bathroom in France.. let's get back to bed in switzerland :D | 07:23 |
| arne_wiebalck | Also, when I go from my office to grab coffee across the street, I cross border countries as the border goes through CERN. | 07:23 |
| uzumaki | arne_wiebalck, that should feel so cool.. of course, over time you'll get used to it | 07:23 |
| arne_wiebalck | uzumaki: you don't realise until someone tells you :-D | 07:23 |
| uzumaki | arne_wiebalck, =D | 07:24 |
| Qianbiao | arne_wiebalck i will prepare my resume now | 07:24 |
| arne_wiebalck | Qianbiao: Depending on your status, there lots of opportunities to come and work at CERN ... and have cross border coffees :) | 07:27 |
| uzumaki | arne_wiebalck, that would be genial!! | 07:27 |
| Qianbiao | :) will consider that. | 07:30 |
| openstackgerrit | wu.shiming proposed openstack/ironic-python-agent-builder master: Bump hacking min version to 3.0.1 https://review.opendev.org/753240 | 07:31 |
| *** dougsz has joined #openstack-ironic | 07:37 | |
| *** rcernin has joined #openstack-ironic | 07:40 | |
| iurygregory | I want another mid cycle at CERN | 07:40 |
| iurygregory | =( | 07:40 |
| *** tosky has joined #openstack-ironic | 07:40 | |
| uzumaki | iurygregory, we all do :D | 07:40 |
| iurygregory | hehehe | 07:41 |
| uzumaki | would be mon plaisir to meet monsieur arne_wiebalck :D | 07:42 |
| iurygregory | this achievement I already have =D | 07:44 |
| arne_wiebalck | iurygregory: uzumaki: once things get back to normal, we can certainly check if we can organise another one here, it was quite fun I think | 07:44 |
| iurygregory | arne_wiebalck, yup! | 07:44 |
| * arne_wiebalck smells fondue | 07:45 | |
| uzumaki | arne_wiebalck, counting on it! | 07:45 |
| * uzumaki wants to try all sorts of French food | 07:45 | |
| uzumaki | iurygregory, lucky duck! :D | 07:45 |
| uzumaki | would be fun to have a French bread quiz :D I don't know many names, but I want to try them all | 07:46 |
| iurygregory | lol | 07:47 |
| uzumaki | :D | 07:48 |
| uzumaki | brioche, croissant, baguette | 07:49 |
| *** patchbot has quit IRC | 08:13 | |
| *** lucasagomes has joined #openstack-ironic | 08:13 | |
| janders | arne_wiebalck uzumaki we can barely cross internal borders here and you're talking living and working across different states :) worlds apart | 08:18 |
| * janders is trying to imagine a police checkpoint in the middle of the corridor between a bedroom and a bathroom | 08:19 | |
| arne_wiebalck | janders: this also comes with complications in this specific situation, e.g.: work, schools, doctors, ... are in a different country and when the borders get close, things get "interesting" | 08:21 |
| janders | arne_wiebalck I can imagine... | 08:22 |
| *** priteau has joined #openstack-ironic | 08:23 | |
| arne_wiebalck | Ironic 101 question: the boot mode is defined as a node property since there is no reliable way to get the boot mode from the node? | 08:23 |
| arne_wiebalck | dtantsur|afk: ^^ | 08:23 |
| * arne_wiebalck assumes just like JayF that dtantsur|afk and TheJulia have the answer to all questions | 08:24 | |
| iurygregory | hey janders o/ | 08:27 |
| *** jtomasek has quit IRC | 08:27 | |
| Qianbiao | Hello, i got an issue when using devstack. | 08:30 |
| iurygregory | python question which style you would prefer http://paste.openstack.org/show/798191/ ? | 08:30 |
| Qianbiao | swift can not restart after machine reboot. | 08:30 |
| Qianbiao | anyone has some issue? | 08:30 |
| iurygregory | wait what?! | 08:31 |
| iurygregory | you did reboot in a machine with devstack? | 08:31 |
| Qianbiao | yes | 08:31 |
| iurygregory | it's gone | 08:31 |
| iurygregory | re deploy | 08:31 |
| Qianbiao | .... | 08:31 |
| Qianbiao | devstack env can not be restart? | 08:31 |
| iurygregory | you can restart services | 08:31 |
| iurygregory | but if you shutdown/reboot the machine with devstack it's over | 08:32 |
| Qianbiao | ...... any reason | 08:32 |
| iurygregory | afaik devstack is something you should just use to do some testing and that's it | 08:32 |
| iurygregory | the goal of devstack doesn't include fault tolerance etc =) | 08:33 |
| Qianbiao | ~~ I have deploy an customer hardware SDN on that env. | 08:34 |
| Qianbiao | if re-deploy devstack, everything need to re-setup | 08:34 |
| iurygregory | yep | 08:34 |
| iurygregory | truth | 08:34 |
| iurygregory | I would copy the configs you need etc and re run | 08:35 |
| *** derekh has joined #openstack-ironic | 08:35 | |
| *** priteau has quit IRC | 08:35 | |
| iurygregory | sorry for the bad news Qianbiao =( | 08:36 |
| Qianbiao | :( | 08:43 |
| *** priteau has joined #openstack-ironic | 08:50 | |
| uzumaki | janders, same is true for Pakistan, but it is just so _interesting_ as arne_wiebalck says, to imagine a country with 'imaginary' borders | 08:55 |
| uzumaki | janders, I'm trying to imagine having to verify my nationality before being permitted to the bathroom =D | 08:56 |
| *** rcernin has quit IRC | 08:59 | |
| *** dtantsur|afk is now known as dtantsur | 09:00 | |
| *** rcernin has joined #openstack-ironic | 09:00 | |
| uzumaki | guten tag dtantsur ! o/ | 09:00 |
| uzumaki | quick question, does automated_cleaning remove RAID volumes? iurygregory dtantsur janders arne_wiebalck | 09:02 |
| arne_wiebalck | uzumaki: software RAID devices, you mean? | 09:02 |
| uzumaki | yes, and hardware RAID volumes as well arne_wiebalck | 09:03 |
| arne_wiebalck | uzumaki: software RAID devices: no, this is not part of automated cleaning | 09:03 |
| uzumaki | I see, and what about hardware RAID volumes? arne_wiebalck | 09:03 |
| arne_wiebalck | uzumaki: and since we followed the way paved by h/w RAID, I don't think auto cleaning would touch h/w RAID devices either | 09:04 |
| uzumaki | I see, thanks arne_wiebalck | 09:04 |
| uzumaki | and do you have any idea what would happen is both software and hardware raid settings are provided for a node? Error state? arne_wiebalck | 09:04 |
| uzumaki | *if | 09:05 |
| arne_wiebalck | in the raid config? | 09:05 |
| uzumaki | yes | 09:05 |
| arne_wiebalck | in the target raid config? | 09:05 |
| dtantsur | arne_wiebalck: boot mode is a wild story. we can get it for some drivers (redfish), we have assert it for others (ipmitool) | 09:06 |
| arne_wiebalck | dtantsur: wild == annoying | 09:06 |
| dtantsur | our boot mode handling is a hybrid of both, and I'm afraid it's only documented in the code | 09:06 |
| arne_wiebalck | dtantsur: in our fleet we have some BMCs which report it properly, others don't | 09:06 |
| dtantsur | https://opendev.org/openstack/ironic/src/branch/master/ironic/drivers/modules/boot_mode_utils.py#L74-L87 | 09:07 |
| arne_wiebalck | uzumaki: if carefully done, in the correct order, ... I'd think this might work for creation :) There was some talk lately what should happen during clean up, though. | 09:08 |
| iurygregory | I have a chicken egg problem for stable/train in IPE ... CI complains about W504.. if I fix locally an run pep8 it complains about W503... http://paste.openstack.org/show/798191/ | 09:08 |
| iurygregory | should we ignore W504? | 09:08 |
| uzumaki | I really wish we had the option to cook them both and eat them =D iurygregory | 09:09 |
| uzumaki | arne_wiebalck, interessant.. I'm wondering if there would be any real world use case for having both types of RAID | 09:09 |
| arne_wiebalck | dtantsur: thanks for the pointer! this reads ... wild :-D | 09:09 |
| * iurygregory didn't get the joke .-. | 09:09 | |
| * uzumaki made a bad joke, it's okay to not get it xD | 09:09 | |
| dtantsur | iurygregory: I think one of them should be ignored, check other repos which | 09:12 |
| iurygregory | dtantsur, ack | 09:12 |
| dtantsur | iurygregory: we seem to ignore W503 | 09:12 |
| * dtantsur has checked ironic-lib | 09:12 | |
| iurygregory | well I know ussuri and master didn't complain for W504... for IPE ofc | 09:13 |
| iurygregory | yeah we do have W503 on ironic also | 09:13 |
| *** Lucas_Gray has joined #openstack-ironic | 09:14 | |
| *** ociuhandu has joined #openstack-ironic | 09:16 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic-prometheus-exporter stable/train: Fallback to `node_uuid` if`node_name` is not present https://review.opendev.org/753067 | 09:16 |
| *** patchbot has joined #openstack-ironic | 09:18 | |
| *** rcernin has quit IRC | 09:19 | |
| openstackgerrit | Ankit Kumar proposed openstack/ironic master: Adding changes for iso less vmedia support https://review.opendev.org/752001 | 09:32 |
| *** k_mouza has joined #openstack-ironic | 09:32 | |
| arne_wiebalck | dtantsur: I think *in theory* you can also get it from ipmitool, the issue is that the BMC does not report it correctly (and ipmitool is only parsing what the BMC returns) | 09:34 |
| arne_wiebalck | *the BMC does not necessarily report it correctly ... | 09:35 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic-inspector bugfix/10.3: Limit inspector jobs to 1 testing VM https://review.opendev.org/753273 | 09:43 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Deprecate the iscsi deploy interface https://review.opendev.org/750204 | 09:46 |
| dtantsur | This is another gate fix we'll need: https://review.opendev.org/#/c/753094/ | 09:48 |
| patchbot | patch 753094 - ironic - Limit inspector jobs to 1 testing VM - 1 patch set | 09:48 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 09:54 |
| iurygregory | dtantsur, ack | 09:54 |
| iurygregory | we need to wait for https://review.opendev.org/#/c/753184 to merge right? | 09:55 |
| patchbot | patch 753184 - ironic - Reduce VMs for multinode and standalone jobs - 1 patch set | 09:55 |
| iurygregory | funny sushy-oem-idrac is not even running victoria tests XD | 09:56 |
| dtantsur | iurygregory: at least the parent (grenade) patch | 09:57 |
| iurygregory | ack | 09:57 |
| Qianbiao | dtantsur several core member has question about https://review.opendev.org/#/c/752024/5 | 10:05 |
| patchbot | patch 752024 - ironic-python-agent - Fix: make Intel CNA hardware manager none generic - 5 patch sets | 10:05 |
| Qianbiao | should we add some spec to clarify the design intent. or need a re-design? | 10:05 |
| dtantsur | Qianbiao: you do what JayF proposed and just disable the LLDP agent in evaluate_hardware_support if LLDP collection is enabled (it's a conf option) | 10:06 |
| Qianbiao | dtantsur not only JayF, include Steve and TheJulia both have question about why we raise exception there. | 10:07 |
| Qianbiao | JayF is talking about a new solution btw. | 10:08 |
| dtantsur | * sorry, "you could do" | 10:12 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic-inspector bugfix/10.3: Fix l-c and mysql GRANT syntex for Ubuntu Focal https://review.opendev.org/753277 | 10:20 |
| Qianbiao | dtantsur ok, that is a solution too. | 10:21 |
| openstackgerrit | Yogesh Ramachandra proposed openstack/ironic master: Fix idrac-wsman bios factory-reset setting https://review.opendev.org/748696 | 10:25 |
| openstackgerrit | Yogesh Ramachandra proposed openstack/ironic master: Fix idrac-wsman bios factory-reset setting https://review.opendev.org/748696 | 10:33 |
| *** k_mouza has quit IRC | 10:35 | |
| openstackgerrit | Merged openstack/ironic master: Reduce grenade node count https://review.opendev.org/753183 | 10:36 |
| *** rcernin has joined #openstack-ironic | 10:37 | |
| *** rcernin has quit IRC | 10:37 | |
| *** rcernin has joined #openstack-ironic | 10:37 | |
| * iurygregory brb lunch | 10:43 | |
| *** k_mouza has joined #openstack-ironic | 10:44 | |
| *** k_mouza has quit IRC | 10:48 | |
| *** k_mouza has joined #openstack-ironic | 10:48 | |
| *** priteau has quit IRC | 11:08 | |
| *** rcernin has quit IRC | 11:11 | |
| openstackgerrit | Merged openstack/bifrost master: Always setup environment in bifrost-cli and make it less verbose https://review.opendev.org/750315 | 11:11 |
| *** bfournie has quit IRC | 11:18 | |
| *** bfournie has joined #openstack-ironic | 11:19 | |
| openstackgerrit | Yogesh Ramachandra proposed openstack/ironic master: Fix idrac-wsman bios factory-reset setting https://review.opendev.org/748696 | 11:28 |
| openstackgerrit | Merged openstack/ironic master: Reduce VMs for multinode and standalone jobs https://review.opendev.org/753184 | 11:41 |
| openstackgerrit | Merged openstack/ironic-python-agent master: Documentation: fix incorrect step names https://review.opendev.org/753080 | 11:44 |
| *** rh-jelabarre has joined #openstack-ironic | 11:57 | |
| *** rh-jelabarre has quit IRC | 11:58 | |
| *** rh-jelabarre has joined #openstack-ironic | 11:58 | |
| openstackgerrit | Merged openstack/sushy master: Make Actions field in Volume resource optional https://review.opendev.org/751900 | 11:59 |
| *** jtomasek has joined #openstack-ironic | 11:59 | |
| openstackgerrit | Ankit Kumar proposed openstack/ironic master: Adding changes for iso less vmedia support https://review.opendev.org/752001 | 12:17 |
| *** jhesketh has quit IRC | 12:32 | |
| *** rcernin has joined #openstack-ironic | 12:55 | |
| *** mgoddard has quit IRC | 13:04 | |
| iurygregory | really CI?! openstack-tox-docs TIMED_OUT .... | 13:07 |
| iurygregory | (╯°□°)╯︵ ┻━┻ | 13:07 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: CI: Lower memory usage of VMs/Increase swap https://review.opendev.org/753057 | 13:09 |
| TheJulia | good morning ^^^ since ci is touchy | 13:09 |
| dtantsur | morning TheJulia | 13:09 |
| iurygregory | good morning TheJulia =) | 13:09 |
| TheJulia | iurygregory: ask in openstack infra if there is any way we can extend that job timeout because we keep encountering it | 13:09 |
| *** mgoddard has joined #openstack-ironic | 13:09 | |
| *** rcernin has quit IRC | 13:10 | |
| iurygregory | it's possible I think | 13:10 |
| iurygregory | let me try some hack here and ask infra =) | 13:10 |
| TheJulia | we're not the only ones who has seen it as well | 13:10 |
| * iurygregory checks on zuul | 13:10 | |
| *** k_mouza has quit IRC | 13:12 | |
| iurygregory | lol the parent jobs doesn't have timeout set | 13:14 |
| *** k_mouza has joined #openstack-ironic | 13:14 | |
| iurygregory | zuul magic for default timeout probably... | 13:14 |
| openstackgerrit | QianBiao Ng proposed openstack/ironic-python-agent master: Fix: make Intel CNA hardware manager none generic https://review.opendev.org/752024 | 13:16 |
| Qianbiao | stevebaker dtantsur ^^ the new changes for CNA card lldp disable process, unittest have not been updated. | 13:18 |
| Qianbiao | JayF | 13:19 |
| Qianbiao | If the code is ok, i will update tests a bit later. | 13:19 |
| *** jhesketh has joined #openstack-ironic | 13:19 | |
| iurygregory | TheJulia, https://review.opendev.org/#/c/752584/ the reason for the timeouts... | 13:21 |
| patchbot | patch 752584 - openstack-zuul-jobs - Unwind the texlive-full addition - 1 patch set | 13:21 |
| TheJulia | wheeeee | 13:22 |
| iurygregory | 3gb of packages Lmaoooooooo | 13:22 |
| dtantsur | Oo | 13:22 |
| iurygregory | to build pdfs | 13:22 |
| iurygregory | (╯°□°)╯︵ ┻━┻ | 13:22 |
| TheJulia | what ever happened to generate some postscript and wrap it in a pdf.... | 13:23 |
| iurygregory | raise your hand if we will put docs non-voting XD | 13:24 |
| TheJulia | can I have a few whiskey's first | 13:24 |
| TheJulia | ? | 13:24 |
| iurygregory | totally | 13:24 |
| dtantsur | iurygregory: I'd be more than open to disabling PDF build for now | 13:25 |
| *** rloo has joined #openstack-ironic | 13:25 | |
| Qianbiao | I guess it maybe use chrome core to render pdf ? | 13:26 |
| Qianbiao | how it could be 3GB... | 13:26 |
| janders | see you tomorrow Ironic o/ | 13:27 |
| iurygregory | dtantsur, put non-voting to unblock CI would be more easy no? (at least on my mind) | 13:27 |
| iurygregory | janders, bye o/ | 13:27 |
| Qianbiao | see you janders | 13:27 |
| TheJulia | goodnight | 13:28 |
| * TheJulia goes and coffeeeeeees | 13:28 | |
| Qianbiao | TheJulia do u like chinese tea. | 13:29 |
| Qianbiao | maybe health than coffee | 13:29 |
| TheJulia | I do actually love good tea | 13:32 |
| TheJulia | but I don't wake up without the caffeine in coffee | 13:32 |
| iurygregory | pro·gram·mer (n) An organism capable of converting caffeine into code. | 13:33 |
| TheJulia | lol | 13:33 |
| iurygregory | that's how it works normally =P | 13:33 |
| iurygregory | TheJulia, this is new for you?! | 13:34 |
| iurygregory | =O | 13:34 |
| dtantsur | iurygregory: broken docs jobs prevent publishing of the docs | 13:37 |
| *** jtomasek has quit IRC | 13:38 | |
| iurygregory | yeah that makes sense | 13:38 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Deprecate the iscsi deploy interface https://review.opendev.org/750204 | 13:39 |
| dtantsur | hopefully this ^^ should finally pass the CI | 13:40 |
| iurygregory | I think tox_skip_pdf would do the job | 13:41 |
| *** Lucas_Gray has quit IRC | 13:43 | |
| *** tzumainn has joined #openstack-ironic | 13:43 | |
| * iurygregory cheks with infra | 13:44 | |
| *** priteau has joined #openstack-ironic | 13:46 | |
| *** abdysn has quit IRC | 13:47 | |
| *** uzumaki has quit IRC | 13:52 | |
| *** Goneri has joined #openstack-ironic | 13:56 | |
| Qianbiao | TheJulia tea is enough for me, no coffee needed. :) | 14:14 |
| openstackgerrit | Zane Bitter proposed openstack/sushy-tools master: Fix race condition initialising persistent dict https://review.opendev.org/752953 | 14:24 |
| TheJulia | I'm pulling really old photos of mine out for a slide deck.... It will either resonate or fall flat. | 14:30 |
| *** Lucas_Gray has joined #openstack-ironic | 14:31 | |
| *** k_mouza has quit IRC | 14:36 | |
| Qianbiao | Hello, in pxeconfig file, which conf will point to the http ip | 14:39 |
| *** k_mouza has joined #openstack-ironic | 14:43 | |
| *** iurygregory has quit IRC | 14:46 | |
| *** iurygregory has joined #openstack-ironic | 14:51 | |
| rpioso | Good morning, ironic | 14:53 |
| dtantsur | morning rpioso | 14:54 |
| rpioso | dtantsur: :-) | 14:54 |
| iurygregory | morning rpioso | 14:54 |
| rpioso | iurygregory: o/ | 14:55 |
| iurygregory | o/ | 14:55 |
| rpioso | iurygregory: What needs to be done to get sushy-oem-idrac to run Victoria tests? | 14:55 |
| iurygregory | rpioso, https://review.opendev.org/#/c/745527/ | 14:56 |
| patchbot | patch 745527 - x/sushy-oem-idrac - Use victoria tests - 1 patch set | 14:56 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Make the standalone-redfish job voting https://review.opendev.org/753374 | 14:56 |
| *** jawad_axd has quit IRC | 14:56 | |
| * iurygregory facepalm ^ I forgot to put voting.... | 14:57 | |
| rpioso | iurygregory: Thank you! | 14:59 |
| iurygregory | rpioso, yw o/ | 14:59 |
| JayF | TheJulia: iurygregory: I'm going to have to start a drug-free revolution in software development. Until we all stop taking performance enhancing drugs, we're all victims of Big Caffeine :D /s | 15:00 |
| TheJulia | oh my.... | 15:00 |
| * TheJulia sips coffee and tries to hammer out slides | 15:01 | |
| JayF | I've already had about 8 oz of cold brew and have a can of soda beside me :D | 15:02 |
| gmann | iurygregory: Re: on Focal failure, there is no gunicorn3 pckg in Focal - https://packages.ubuntu.com/search?suite=focal&keywords=gunicorn3 | 15:04 |
| *** rcernin has joined #openstack-ironic | 15:06 | |
| clarkb | gmann: https://pypi.org/project/gunicorn/ you can install it from pypi though | 15:07 |
| dtantsur | TheJulia: do I get it right that we need stable/victoria soon? | 15:09 |
| TheJulia | relatively soon would be preferred | 15:10 |
| dtantsur | do we have volunteers for the pre-release dance (release mapping, reno cleanup, etc)? | 15:10 |
| *** rcernin has quit IRC | 15:10 | |
| dtantsur | TheJulia: I guess most projects will have it by the end of this week? | 15:10 |
| dtantsur | meaning, if we don't want troubles with grenade, we need it too | 15:10 |
| TheJulia | dtantsur: yeah, we can always temporarily disable grenade but they improved the logic... at least in the old grenade jobs and changed it in R-2 last cycle if memory serves | 15:11 |
| dtantsur | cool. still, requirements branching and stuff.. we shouldn't take too long with it | 15:11 |
| TheJulia | yeah | 15:11 |
| openstackgerrit | Ghanshyam Mann proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 15:11 |
| gmann | clarkb: yeah | 15:12 |
| gmann | iurygregory: ^^ let's see now | 15:12 |
| dtantsur | TheJulia: I guess the iscsi deprecation is blocking a release on the ironic side, IPA TLS - on the IPA side, and inspector should be fine? | 15:12 |
| TheJulia | the reno stuff was on my list for this week, I just REALLY need to get this presentation into a state where it can be practiced and then recorded soon :( | 15:12 |
| TheJulia | dtantsur: indeed | 15:12 |
| iurygregory | gmann, tks for updating I was in a meeting | 15:12 |
| TheJulia | I was actually going to ask you if your good with calling inspector done for the cycle | 15:12 |
| dtantsur | I'll double-check right now | 15:13 |
| iurygregory | s/was/am | 15:13 |
| TheJulia | I've got a bugfix but it can always be backported | 15:13 |
| dtantsur | https://review.opendev.org/#/c/751660/ is a feature, and we need to fix the CI to merge it | 15:13 |
| patchbot | patch 751660 - ironic-inspector - Add NVIDIA V100 to known accelerator devices - 1 patch set | 15:13 |
| dtantsur | bifrost also has a few outstanding patches as well | 15:14 |
| dtantsur | https://review.opendev.org/#/c/749619/ looks like an awesome last-minute feature | 15:15 |
| patchbot | patch 749619 - ironic - Redfish driver firmware update - 19 patch sets | 15:15 |
| openstackgerrit | Merged x/sushy-oem-idrac master: Use victoria tests https://review.opendev.org/745527 | 15:17 |
| TheJulia | yes, it looks fairly straight forward.... fairly | 15:18 |
| *** belmoreira has quit IRC | 15:21 | |
| *** Lucas_Gray has quit IRC | 15:23 | |
| dtantsur | straightforward, but large | 15:23 |
| * dtantsur is half way through | 15:23 | |
| TheJulia | yeah | 15:24 |
| TheJulia | the testing is what I need lots of coffee for | 15:24 |
| arne_wiebalck | would be great if this could be done on active nodes ... we did have a spec/rfe for this at some point, no? | 15:25 |
| iurygregory | I've looked at a few bifrost patches I think I've +2 already | 15:25 |
| *** Lucas_Gray has joined #openstack-ironic | 15:26 | |
| TheJulia | arne_wiebalck: we talked about making a state to formally do it but never really reached a final consensus on mechanics. Maybe we should devote more time in the PTG to it, Perhaps even schedule more time during the ptg? | 15:26 |
| dtantsur | "active steps" :) | 15:27 |
| dtantsur | PTG++ | 15:27 |
| arne_wiebalck | TheJulia: I was thinking that if we have the howto now, it may be time to revisit this and see how this could be done on nodes which are in production (which I guess would be an even larger target use case). | 15:28 |
| dtantsur | the last time we said "no until we finish with deploy steps". we've largely finished with deploy steps (except for the API bit) | 15:28 |
| arne_wiebalck | ptg sounds good :) | 15:28 |
| JayF | Assuming you all are talking about inspection on active nodes, I think we should in general make more things doable on active nodes. Including manual cleaning. | 15:34 |
| TheJulia | well, we already have a flag to enable that via admin self submission if that maeks sense | 15:35 |
| TheJulia | but yeah, admiistrative actions/capabilities are really needed for running/deployed nodes | 15:35 |
| TheJulia | likely with explicit policy permissions | 15:35 |
| JayF | I don't know what admin self-submissino means, but it's sweet if you can active->cleaning->active with manual cleaning. | 15:36 |
| JayF | That would enable some good use cases for Ironic being the one-stop-shop for firmware management | 15:36 |
| TheJulia | admin self submission for introspection data | 15:36 |
| dtantsur | even with booting into IPA? | 15:36 |
| openstackgerrit | Richard G. Pioso proposed x/sushy-oem-idrac master: Fix GET HTTP 202 status without Location https://review.opendev.org/726275 | 15:36 |
| iurygregory | JayF, please don't take my coffee away D: (I do miss good coffee like brazilian coffee) | 15:39 |
| JayF | I'm not taking your coffee away :D | 15:40 |
| openstackgerrit | QianBiao Ng proposed openstack/ironic-python-agent master: Fix: make Intel CNA hardware manager none generic https://review.opendev.org/752024 | 15:43 |
| Qianbiao | JayF dtantsur unittest updated, should be good to go now. | 15:44 |
| dtantsur | iurygregory: do you plan on disabling PDF build in the docs? I see another failure.. | 15:49 |
| iurygregory | dtantsur, ouch, I will push the patch | 15:50 |
| iurygregory | only ironic? | 15:50 |
| dtantsur | I've only seen ironic affected | 15:50 |
| iurygregory | ack | 15:50 |
| openstackgerrit | QianBiao Ng proposed openstack/ironic-python-agent master: Fix: make Intel CNA hardware manager none generic https://review.opendev.org/752024 | 15:52 |
| *** ociuhandu_ has joined #openstack-ironic | 15:55 | |
| *** bdodd has quit IRC | 15:55 | |
| *** bdodd has joined #openstack-ironic | 15:56 | |
| *** ociuhandu has quit IRC | 15:58 | |
| *** lucasagomes has quit IRC | 15:58 | |
| *** ociuhandu_ has quit IRC | 16:00 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: Disable pdf builds https://review.opendev.org/753396 | 16:07 |
| *** irclogbot_3 has quit IRC | 16:08 | |
| iurygregory | LOL Fujitsu iRMC CI reports merge failed in something I just submited lmao | 16:08 |
| *** irclogbot_0 has joined #openstack-ironic | 16:09 | |
| ajya | iurygregory: earlier today it started to do it in other patches too | 16:09 |
| iurygregory | funny message tbh XD | 16:10 |
| ajya | iurygregory: any idea why that's happening? Though it seems it is safe to ignore it. | 16:11 |
| iurygregory | ajya, safe to ignore =) probably they are trying to restore their 3rd Party CI =) | 16:11 |
| ajya | that was my guess too | 16:12 |
| dtantsur | that's correct. I guess something in the setup is not settled yet | 16:18 |
| TheJulia | Yeah, they emailed me and indicated physical access to the machines has been severely hampering efforts to get everything in the physical deployment right to get CI running | 16:22 |
| TheJulia | Becasue... pandemic! | 16:22 |
| *** Qianbiao has quit IRC | 16:23 | |
| TheJulia | would anyone like to volunteer a favorate data center photo that I could use in a presentation? Hoping for racks of serers | 16:25 |
| TheJulia | The more the better :) | 16:25 |
| *** k_mouza has quit IRC | 16:25 | |
| *** johnsom has quit IRC | 16:26 | |
| *** johnsom has joined #openstack-ironic | 16:27 | |
| jroll | TheJulia: asking around here for you :) | 16:30 |
| TheJulia | trandles: looks like you guys have a few photos on flikr that are creative commons | 16:31 |
| TheJulia | trandles: https://www.flickr.com/photos/losalamosnatlab/5907942748/in/album-72157648722396370/ could work | 16:32 |
| TheJulia | sadly roadrunner photos are all rights reserved | 16:32 |
| TheJulia | :( | 16:32 |
| *** dougsz has quit IRC | 16:33 | |
| *** k_mouza has joined #openstack-ironic | 16:34 | |
| *** dtantsur is now known as dtantsur|afk | 16:34 | |
| dtantsur|afk | o/ | 16:34 |
| trandles | huh, we have a flickr presence? | 16:36 |
| TheJulia | trandles: apparently so! | 16:37 |
| trandles | a lot of that is old | 16:37 |
| trandles | hahahahaha the wall between the data center and my office was immediately behind the photographer for this | 16:38 |
| trandles | https://www.flickr.com/photos/losalamosnatlab/16511465605/in/album-72157648722396370/ | 16:38 |
| trandles | it was noisy at times | 16:38 |
| trandles | that's Trinity water cooling construction | 16:38 |
| *** k_mouza has quit IRC | 16:38 | |
| TheJulia | so that could actually work | 16:38 |
| TheJulia | because they are building a foundation, and what is bare metal for computing, a substrate or foundation | 16:39 |
| trandles | Crayons installing Trinity in this pic | 16:39 |
| trandles | https://www.flickr.com/photos/losalamosnatlab/18785323579/in/album-72157648722396370/ | 16:39 |
| openstackgerrit | Richard G. Pioso proposed openstack/ironic master: Fix redfish BIOS to use @Redfish.SettingsApplyTime https://review.opendev.org/752614 | 16:41 |
| *** k_mouza has joined #openstack-ironic | 16:42 | |
| * rpioso bows to the IT ... and reboots his system | 16:43 | |
| *** k_mouza has quit IRC | 16:46 | |
| *** k_mouza has joined #openstack-ironic | 16:51 | |
| *** derekh has quit IRC | 16:58 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent master: Simplify heartbeating by removing use of select() https://review.opendev.org/746774 | 16:59 |
| *** Lucas_Gray has quit IRC | 17:00 | |
| iurygregory | TheJulia, I will +2+W on https://review.opendev.org/753057 since it's green and we are trying to get CI happy | 17:01 |
| patchbot | patch 753057 - ironic-python-agent - CI: Lower memory usage of VMs/Increase swap - 3 patch sets | 17:01 |
| jroll | TheJulia: https://imgur.com/a/VHpwGrB courtesy of penick | 17:04 |
| iurygregory | jroll, WOW | 17:08 |
| larsks | If ironic inspector seems to be responding to requests from unknown nodes, (a) is that expected, and (b) is there a config option that controls that behavior? | 17:08 |
| iurygregory | looks amazing | 17:08 |
| jroll | :D | 17:08 |
| *** k_mouza has quit IRC | 17:11 | |
| larsks | The problem I'm looking at is that we have some nodes with two connections on the same network. Once provisioned they will be a LACP bond, but during inspection they are both pulling an address even though one MAC is registered, and when both configured the inspection agent is unable to contact the ironic api. | 17:12 |
| larsks | Does that sound familiar to anyone? | 17:12 |
| *** k_mouza has joined #openstack-ironic | 17:18 | |
| *** k_mouza has quit IRC | 17:23 | |
| *** cdearborn has joined #openstack-ironic | 17:28 | |
| * TheJulia reads | 17:29 | |
| TheJulia | iurygregory: awesome | 17:29 |
| TheJulia | larsks: expected behavior if the dnsmasq or iptables filtering is not in use. | 17:31 |
| TheJulia | larsks: I think if you load both macs in inspector will do the needful and block access to the dnsmasq instance | 17:32 |
| larsks | TheJulia: where is that -- either the dnsmasq or iptables filtering -- controlled? | 17:32 |
| openstackgerrit | Ghanshyam Mann proposed openstack/ironic master: migrate testing to ubuntu focal https://review.opendev.org/739448 | 17:33 |
| TheJulia | ironic-inspector service controls it | 17:33 |
| TheJulia | looking up the config options | 17:33 |
| TheJulia | larsks: pxe_filter | 17:34 |
| * larsks looks at the inspector.conf... | 17:34 | |
| larsks | TheJulia: so that's currently set to 'dnsmasq', but I'm not sure what that's supposed to do. The issue is that when triggering an inspection on a *specific* host, the ',ignore' is removed from the unknown_hosts_filter. | 17:35 |
| larsks | So while inspection is active, the inspector will respond to everything. | 17:36 |
| TheJulia | well, you said your only registering one mac right? | 17:36 |
| larsks | Right, yes. | 17:36 |
| TheJulia | so yeah, it bases the config on the macs known if additional macs are not added then I don't think it will pick them up and block access | 17:36 |
| larsks | I'm not following that last statement. With both mac addresses register, does it just enable introspection for one of them? And explicitly block the other? | 17:37 |
| *** k_mouza has joined #openstack-ironic | 17:37 | |
| TheJulia | I think it enables it for both macs and then blocks both | 17:37 |
| larsks | Right, so what we're trying to do is simply not have the inspector ever respond to unknown mac addresses. | 17:38 |
| larsks | Is that sensible? | 17:38 |
| TheJulia | it is sensible with not doing any sort of discovery | 17:38 |
| *** ociuhandu has joined #openstack-ironic | 17:38 | |
| *** jtriley has joined #openstack-ironic | 17:39 | |
| larsks | That's fine then. We're not trying to discover new nodes; we just want inspection to run on nodes that are already registered. | 17:39 |
| TheJulia | so I _think_ the path is to completley nuke inspector's dnsmasq service | 17:39 |
| *** ociuhandu has quit IRC | 17:39 | |
| TheJulia | managed introspection in the conductor should create PXE configs that point to inspector, if memory serves | 17:39 |
| *** ociuhandu has joined #openstack-ironic | 17:40 | |
| larsks | Is there a config setting that controls the behavior w/r/t unknown hosts? Is that "enable_node_discovery"? I don't think that's been explicitly configured... | 17:40 |
| larsks | (and the source suggests it defaults to false) | 17:40 |
| TheJulia | larsks: there is not | 17:40 |
| TheJulia | it relies upon dnsmasq sending unknowns responses to a pxe config for inspector | 17:41 |
| *** k_mouza has quit IRC | 17:41 | |
| TheJulia | enable_node_discovery, when set to true, will create unknown nodes in ironic | 17:41 |
| TheJulia | but of course, it has to boot them to do that | 17:41 |
| TheJulia | you don't even want them getting that far at all | 17:41 |
| TheJulia | iurygregory: I just issued a recheck for https://review.opendev.org/#/c/753081 which should be happier now | 17:42 |
| patchbot | patch 753081 - ironic - CI: Remove the build check for pre-build ramdisks ... - 1 patch set | 17:42 |
| larsks | TheJulia: I'm so sorry, I'm not tracking what you're saying. Is there a way to stop inspector for configuring dnsmasq to respond to unknown hosts during inspection? | 17:42 |
| TheJulia | tell inspector in that case no | 17:42 |
| TheJulia | inspector only configures the filters after the fact | 17:42 |
| TheJulia | the default dnsmasq config is what your needing | 17:42 |
| larsks | Are you saying we need to modify the dnsmasq configuration manually? I thought that was all generated by ironic. | 17:43 |
| iurygregory | TheJulia, ack I will look after dinner and some exercises | 17:43 |
| iurygregory | =) | 17:43 |
| larsks | We've tried expicitly editing the unknown_hosts_filter, but those changes just get overwritten by ironic. | 17:43 |
| *** ociuhandu has quit IRC | 17:45 | |
| TheJulia | unknown_hosts_filter?!? | 17:45 |
| TheJulia | hmm | 17:45 |
| TheJulia | hmmmm | 17:45 |
| TheJulia | looks like harald added support for this in 2018 | 17:46 |
| *** k_mouza has joined #openstack-ironic | 17:46 | |
| jtriley | $ cat /var/lib/ironic-inspector/dhcp-hostsdir/unknown_hosts_filter | 17:47 |
| jtriley | *:*:*:*:*:* | 17:47 |
| jtriley | that thing ^ | 17:47 |
| TheJulia | larsks: what is the pxe_filter config in that deployment? | 17:47 |
| TheJulia | is node_not_found_hook set at all? | 17:48 |
| jtriley | when we're not introspecting that has an ';ignore' appended to the end. as soon as we start introspection it removes it which seems to be what allows the other macs/interfaces that weren't registered to get an IP | 17:48 |
| TheJulia | is the node_not_found_hook set? | 17:49 |
| TheJulia | https://github.com/openstack/ironic-inspector/blob/2c5cf4e6677386cb182f00444e0b3ed248e97923/ironic_inspector/pxe_filter/dnsmasq.py#L49 | 17:50 |
| *** k_mouza has quit IRC | 17:51 | |
| larsks | TheJulia: so if it's not clear, I'm working with jtriley :) | 17:52 |
| TheJulia | Yeah | 17:53 |
| TheJulia | :) | 17:53 |
| jtriley | TheJulia: hi - sorry should have mentioned that earlier :) | 17:53 |
| TheJulia | steppinga way for a few minutes | 17:53 |
| *** tbonds has joined #openstack-ironic | 17:56 | |
| TheJulia | okay, I'm back | 18:06 |
| TheJulia | the reason I'm asking about the node not found hook is because if it is set it looks like, and this is just reading the dnsmasq pxe filter code at a glance, that once inspection is turned on, if populated with any value, it looks like it would change the default filter | 18:08 |
| *** belmoreira has joined #openstack-ironic | 18:13 | |
| openstackgerrit | Richard G. Pioso proposed x/sushy-oem-idrac master: Export System Configuration https://review.opendev.org/710327 | 18:25 |
| *** gyee has joined #openstack-ironic | 18:34 | |
| jtriley | TheJulia: we just checked and it's not set: "#node_not_found_hook = <None>" | 18:39 |
| jtriley | TheJulia: what would you recommend setting that to? we found "enroll" as an option but that's opposite what we'd want in that case it seems | 18:39 |
| hjensas | jtriley: mm, so if memory serves me right there should be an option to opening/not-opening the unknown_host_filter. To allow discovery i.e node_not_found_hook it must be opened. | 18:42 |
| hjensas | jtriley: if you ironic node have a port associated it should only open for that mac addres, and not the unknown_host_filter. | 18:42 |
| jtriley | right, so out of the box without node discovery explicitly enabled (and it seems it defaults to off) it seems the unknown_hosts_filter is being used and allowing any node to get an ip which is the problem we're trying to solve currently | 18:43 |
| hjensas | jtriley: ok, let me look at the code again. It's been a while ... | 18:44 |
| jtriley | we're using LACP on two nics and the issue is during introspection both nics are DHCP'ing and receiving an IP on the provisioning network. we can probably solve this by removing indepdence of those LACP nics at the switch level but the current out-of-the-box behavior seems odd | 18:44 |
| jtriley | in any case thanks a lot to both of you for your help with this | 18:45 |
| openstackgerrit | Merged openstack/ironic-python-agent master: CI: Lower memory usage of VMs/Increase swap https://review.opendev.org/753057 | 18:46 |
| hjensas | jtriley: So I think we may have a bug. It is opening for unknown_hosts when intrispection is active. | 18:53 |
| hjensas | jtriley: the node_not_found_hook will open it period, as any node booting should be "discovered" and added to the inventory. Once discovered the node is known, and it is filtered unless being introspected. | 18:54 |
| tbonds | hjensas: working with jtriley here - the function (_should_enable_unknown_hosts()) description doesn't match the function | 18:54 |
| hjensas | tbonds: yes, that's what my conclusion is too. | 18:55 |
| jtriley | hjensas: ok thanks that was our hunch | 18:56 |
| jtriley | hjensas: is a patch to _should_enable_unknown_hosts the fix in this case? happy to test one out or work on that | 18:56 |
| *** JamesBenson has quit IRC | 18:57 | |
| hjensas | jtriley: I'm looking at the https://opendev.org/openstack/ironic-inspector/src/branch/master/ironic_inspector/pxe_filter/iptables.py atm. | 18:58 |
| hjensas | jtriley: I think we should consider this an RFE, it worked the same way with iptables. | 19:00 |
| hjensas | jtriley: We support adding a node in ironic, with no ports. The send that to introspection, i.e a node is active for introspection but we do not know it's mac yet. So we need to open DHCP for all. | 19:01 |
| hjensas | jtriley: once introspected the ports for discovered interfaces on the node are added to ironic. | 19:01 |
| * hjensas wonder if there should be a config option, 'allow_only_known_macs' or something. | 19:02 | |
| TheJulia | jtriley: fwiw, re lacp, that causes nothing but problems depending on switches, we recently added high level docs on guidance regarding lacp usage but needless to say it is not recommended in general use cases. | 19:03 |
| jtriley | TheJulia: :( hmmmm ok we'll keep that in mind - probably will try it out and abandon if it's a nightmare | 19:05 |
| jtriley | hjensas: agreed and that should probably be the default setting, no? | 19:06 |
| TheJulia | jtriley: it all has to do with switches and settings, fwiw | 19:06 |
| openstackgerrit | Harald Jensås proposed openstack/ironic-inspector master: Correct the docstring for _should_enable_unknown_hosts https://review.opendev.org/753430 | 19:06 |
| TheJulia | jtriley: passive lacp is the best path to go but in that case the deployed OS needs to know to start speaking LACP to bring up the bond. | 19:06 |
| *** zzzeek has quit IRC | 19:08 | |
| TheJulia | jtriley: https://docs.openstack.org/ironic/latest/admin/troubleshooting.html#why-does-x-issue-occur-when-i-am-using-lacp-bonding-with-ipxe | 19:08 |
| *** zzzeek has joined #openstack-ironic | 19:09 | |
| hjensas | I'm not convinced the LACP issue would be persistently solved by not opening the unknown_macs_filter. It would for the first introspection if only one LACP member port is created in ironic, but for the second inspector might have created ironic ports for both members will be present, and we'd end up adding both to the allowed list. | 19:10 |
| jtriley | TheJulia: we have LACP fallback so introspection image doesn't need to setup the bond | 19:11 |
| jtriley | TheJulia: oh yeah you mean the provisioned OS - yeah that would need to known to setup the bond. Reading around it seemed that was supported though, no? | 19:12 |
| TheJulia | jtriley: the provisioned OS needs to be informed, and fallback may still move ports to a transient blocking state, again, depending on switch config/firmware | 19:13 |
| jtriley | hjensas: on the introspection side you could still hit this issue without lacp - just two nics that are trunked would be enough to recreate this problem, no? | 19:16 |
| hjensas | Why is it a problem that both nic's DHCP and recive an address? | 19:19 |
| *** priteau has quit IRC | 19:23 | |
| jtriley | hjensas: so in the past having two nics on the same network can cause problems (kernel might receive traffic on one interface and send out the other) | 19:26 |
| jtriley | hjensas: that said we might hvae just found a different reason the introspection is failing - it seems it can't find the root disk on these nodes | 19:26 |
| jtriley | hjensas: do you know if that would fail the introspection or just report back that there are no disks in the host? | 19:26 |
| hjensas | jtriley: mm, yes the reverse path filter. (But I think that would only happen if LACP was active switch side.) | 19:28 |
| hjensas | jtriley: not sure on the root disk question. | 19:28 |
| TheJulia | re two nics on same network, That would be post deployment, we see people who do that for deployment purposes all the time with the deployment ramdisks | 19:29 |
| jtriley | TheJulia: ok good to know so perhaps our introspections are failing for the disk issue then - we're debugging that now | 19:34 |
| jtriley | in any case it still seems odd that ironic hands out ips to any nic/mac that asks for one without being explicitly told to do so via configuration so I'd still say we have a bug or RFE here | 19:34 |
| jtriley | but I'm also an openstack newb so what do I know :P | 19:35 |
| openstackgerrit | Harald Jensås proposed openstack/ironic-inspector master: dnsmasq pxe filter - option to always block unknown hosts https://review.opendev.org/753435 | 19:41 |
| hjensas | jtriley: TheJulia: there is a patch ^^ adding an option to keep the unknow_filter shut, period. | 19:42 |
| *** rloo has quit IRC | 19:43 | |
| *** rloo has joined #openstack-ironic | 19:44 | |
| * hjensas guess we could also think about only opening unknown_filter in case the node being introspected does not already have a pxe port. | 19:45 | |
| *** jtriley has quit IRC | 19:50 | |
| *** bdodd has quit IRC | 19:54 | |
| *** jtriley has joined #openstack-ironic | 20:01 | |
| jtriley | hjensas: awesome, thanks! we'll try to test that out - guessing we can just bind mount that into the podman container some way | 20:01 |
| hjensas | jtriley: podman mount 'inspector_container' - patch the file, and restart the service. | 20:02 |
| openstackgerrit | Richard G. Pioso proposed x/sushy-oem-idrac master: DRAC: Redfish management clean steps https://review.opendev.org/721596 | 20:03 |
| jtriley | hjensas: awesome, thank you! will let you know how it goes | 20:05 |
| *** ianychoi_ has joined #openstack-ironic | 20:23 | |
| *** ianychoi has quit IRC | 20:27 | |
| *** jtriley has quit IRC | 20:33 | |
| *** ociuhandu has joined #openstack-ironic | 20:34 | |
| *** belmoreira has quit IRC | 20:40 | |
| *** tosky_ has joined #openstack-ironic | 21:09 | |
| *** k_mouza has joined #openstack-ironic | 21:09 | |
| *** tosky is now known as Guest33200 | 21:10 | |
| *** tosky_ is now known as tosky | 21:10 | |
| *** k_mouza has quit IRC | 21:14 | |
| openstackgerrit | Steve Baker proposed openstack/ironic master: New argument validate decorator https://review.opendev.org/742795 | 21:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Add status_code argument to @method.expose https://review.opendev.org/743677 | 21:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Add @method.body decorator https://review.opendev.org/743678 | 21:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Duplicate trait validation with jsonschema https://review.opendev.org/745778 | 21:21 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Utility functions for REST API JSON handling https://review.opendev.org/751160 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert allocations endpoint to plain JSON https://review.opendev.org/743679 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert bios endpoint to plain JSON https://review.opendev.org/745475 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert chassis endpoint to plain JSON https://review.opendev.org/745476 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert conductors endpoint to plain JSON https://review.opendev.org/745779 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert deploy_templates endpoint to plain JSON https://review.opendev.org/745780 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert drivers endpoint to plain JSON https://review.opendev.org/741865 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert event endpoint to plain JSON https://review.opendev.org/746623 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert nodes endpoint to plain JSON https://review.opendev.org/747820 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert ports endpoint to plain JSON https://review.opendev.org/750120 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert portgroups endpoint to plain JSON https://review.opendev.org/750864 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert ramdisk endpoint to plain JSON https://review.opendev.org/750857 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert volume endpoint to plain JSON https://review.opendev.org/750858 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert volume/connectors endpoint to plain JSON https://review.opendev.org/750859 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Convert volume/targets endpoint to plain JSON https://review.opendev.org/750860 | 21:22 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Expunge the internal version of WSME https://review.opendev.org/750861 | 21:22 |
| openstackgerrit | Merged openstack/ironic master: CI: Remove the build check for pre-build ramdisks only https://review.opendev.org/753081 | 21:23 |
| openstackgerrit | Christopher Dearborn proposed openstack/ironic master: Redfish driver firmware update https://review.opendev.org/749619 | 21:36 |
| openstackgerrit | Merged x/sushy-oem-idrac master: Fix GET HTTP 202 status without Location https://review.opendev.org/726275 | 21:58 |
| openstackgerrit | Richard G. Pioso proposed x/sushy-oem-idrac master: DRAC: Redfish management clean steps https://review.opendev.org/721596 | 22:02 |
| *** gmann has quit IRC | 22:04 | |
| *** arne_wiebalck has quit IRC | 22:04 | |
| *** gregwork has quit IRC | 22:04 | |
| *** andrein has quit IRC | 22:04 | |
| *** seongsoocho has quit IRC | 22:04 | |
| *** m1p has quit IRC | 22:04 | |
| *** m1p has joined #openstack-ironic | 22:05 | |
| *** arne_wiebalck has joined #openstack-ironic | 22:05 | |
| *** andrein has joined #openstack-ironic | 22:05 | |
| *** gmann has joined #openstack-ironic | 22:05 | |
| *** seongsoocho has joined #openstack-ironic | 22:05 | |
| *** mnaser has quit IRC | 22:06 | |
| *** gregwork has joined #openstack-ironic | 22:08 | |
| *** rloo has quit IRC | 22:09 | |
| *** mnaser has joined #openstack-ironic | 22:10 | |
| cdearborn | hey folks, I'm struggling to understand the gate failures on https://review.opendev.org/#/c/748927/. settings is clearly defined. I downloaded the patch and pep8 and the unit test pass with no issues. Tried recheck, but the gate failed with the same errors. Any ideas? | 22:11 |
| patchbot | patch 748927 - sushy - Make message parsing more resilient - 3 patch sets | 22:11 |
| clarkb | cdearborn: https://opendev.org/openstack/sushy/src/branch/master/sushy/tests/unit/resources/registry/test_message_registry.py#L22 its been renamed. Zuul tests what the change will look like if it merged to the target branch rather than as provided | 22:15 |
| clarkb | (and this is the reason for that because if we merged that change as is sushy would be broken afterwards) | 22:16 |
| openstackgerrit | Steve Baker proposed openstack/ironic master: Use tempdirs for /tftpboot, /httpboot creation tests https://review.opendev.org/753457 | 22:16 |
| clarkb | really it is a merge conflict that simple text based diffing can't catch, but testing can | 22:16 |
| cdearborn | clarkb, thanks much! that's a really weird one. wish it would just say there's a merge conflict. will update | 22:17 |
| clarkb | cdearborn: well git can't detect that merge conflict is the issue | 22:18 |
| clarkb | but our testing has and did | 22:18 |
| clarkb | but the testing doesn't know if it is due to the merge or a typo | 22:19 |
| openstackgerrit | Christopher Dearborn proposed openstack/sushy master: Make message parsing more resilient https://review.opendev.org/748927 | 22:33 |
| *** ociuhandu has quit IRC | 22:46 | |
| *** tosky has quit IRC | 22:48 | |
| *** rcernin has joined #openstack-ironic | 22:55 | |
| *** tkajinam has joined #openstack-ironic | 22:57 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent bugfix/6.3: CI: Lower memory usage of VMs/Increase swap https://review.opendev.org/753460 | 23:01 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent bugfix/6.2: CI: Lower memory usage of VMs/Increase swap https://review.opendev.org/753461 | 23:01 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent stable/ussuri: CI: Lower memory usage of VMs/Increase swap https://review.opendev.org/753462 | 23:01 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-inspector bugfix/10.2: Limit inspector jobs to 1 testing VM https://review.opendev.org/753463 | 23:14 |
| openstackgerrit | Julia Kreger proposed openstack/ironic bugfix/15.2: CI: Remove the build check for pre-build ramdisks only https://review.opendev.org/753466 | 23:21 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/ussuri: CI: Remove the build check for pre-build ramdisks only https://review.opendev.org/753467 | 23:22 |
| clarkb | fungi thinks that https://review.opendev.org/752584 will address the pdf doc timeouts (it fixes nova) | 23:23 |
| patchbot | patch 752584 - openstack-zuul-jobs - Replace texlive-full with cm-super (MERGED) - 2 patch sets | 23:23 |
| TheJulia | clarkb: awesome, blocking disable patch | 23:23 |
| fungi | yeah, i did not test ironic's docs, but nova's went from frequently hitting the 20 minute timeout to completing in 10 on my dnm canary change | 23:26 |
| openstackgerrit | Julia Kreger proposed openstack/ironic bugfix/15.2: CI: Patch to fix CI jobs in bugfix/stable branches https://review.opendev.org/753468 | 23:29 |
| *** zzzeek has quit IRC | 23:30 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic bugfix/15.1: CI: Patch to fix CI jobs in bugfix/stable branches https://review.opendev.org/753469 | 23:31 |
| *** zzzeek has joined #openstack-ironic | 23:31 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/ussuri: CI: Patch to fix CI jobs in bugfix/stable branches https://review.opendev.org/753470 | 23:34 |
| TheJulia | fungi: that is a super good sign | 23:35 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/train: CI: Patch to fix CI jobs in bugfix/stable branches https://review.opendev.org/753471 | 23:43 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!