| vanou | good morning ironic | 04:49 |
|---|---|---|
| *** akahat|rover is now known as akahat | 07:05 | |
| dtantsur | kubajj: re Red Hat: the office in Brno is hiring nearly constantly. I know you're not really fond of Brno though ;) | 09:30 |
| rpittau | good morning ironic! o/ | 09:37 |
| kubajj | Morning everyone! | 11:01 |
| kubajj | dtantsur: I've seen some remote jobs on the website, so I think I could work from Prague. But I think I still have time to figure stuff out. | 11:03 |
| dtantsur | true | 11:03 |
| kubajj | dtantsur: do you agree that if swift object (either inventory or plugin) exists, but fails to delete, we just log it and let the node delete? | 11:04 |
| opendevreview | Merged openstack/ironic master: Imported Translations from Zanata https://review.opendev.org/c/openstack/ironic/+/872060 | 11:11 |
| dtantsur | kubajj: I'm largely undecided. That was what I suggested, but I think TheJulia objected? Or do I remember it wrong? | 11:11 |
| iurygregory | good morning Ironic | 11:16 |
| kubajj | dtantsur: I think TheJulia doesn't like to orphan things, and JayF suggested we orphan and make loud noises about orphaning | 11:16 |
| dtantsur | kubajj: option #3 which we already use in some cases: only allow orphaning in maintenance mode | 11:22 |
| dtantsur | We do this because logs are not really visible for e.g. a CLI user | 11:31 |
| kubajj | dtantsur: when is maintainance used? | 11:50 |
| dtantsur | kubajj: it's a boolean flag on a node that lets operators mark a node as "undergoing maintenance" or not operating properly | 12:17 |
| dtantsur | when this mode is on, we allow actions with such a node that are not allowed normally | 12:17 |
| dtantsur | like forced deletion | 12:17 |
| kubajj | dtantsur: would anyone destroy a node without maintenance then? | 12:18 |
| dtantsur | kubajj: yeah, it's normal to delete nodes that are not in maintenance. but Ironic is more picky about this. | 12:19 |
| dtantsur | e.g. it will let you delete nodes in "available" and "manageable" states, but not "active" | 12:19 |
| kamleshChauvhan | Hi Ironic. | 13:11 |
| *** tkajinam is now known as Guest2882 | 13:11 | |
| kamleshChauvhan | can someone please help to second review for backported patch https://review.opendev.org/c/openstack/sushy/+/868777 | 13:11 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent bugfix/8.3: Fix CI https://review.opendev.org/c/openstack/ironic-python-agent/+/871980 | 13:39 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent bugfix/8.3: Fix CI https://review.opendev.org/c/openstack/ironic-python-agent/+/871980 | 13:42 |
| TheJulia | dtantsur: ahh, good idea w/r/t maintenance | 14:14 |
| dtantsur | good morning TheJulia | 14:22 |
| * TheJulia attempts to wake up | 14:25 | |
| opendevreview | Jakub Jelinek proposed openstack/ironic master: Erase swift inventory entry on node deletion https://review.opendev.org/c/openstack/ironic/+/871394 | 14:50 |
| kubajj | dtantsur, or TheJulia: could either of you see why test_inspect_utils.SwiftCleanUp.test_clean_up_swift_entries_with_exception throws the SwiftObjectStillExists exception even though I set http_status to 404 in the change above? | 14:51 |
| TheJulia | Would anyone be willing to review the sharding patches? We're starting to run short on time and we need them merged to finally begin fixing the issue with nova | 14:55 |
| TheJulia | kubajj: weeeird | 14:55 |
| TheJulia | kubajj: line 96 should be return, not pass | 14:56 |
| TheJulia | sorry! | 14:56 |
| kubajj | TheJulia: I see. Thanks | 14:56 |
| JayF | TheJulia: fwiw I'm going to focus on the port querying part of that today | 14:58 |
| JayF | TheJulia: explicitly at the expense of almost anything else | 14:58 |
| JayF | so someone else is gonna have to guard the gates lol | 14:58 |
| * TheJulia is sad she never got a guard the gate pin | 14:59 | |
| JayF | this is OSS; you want a pin you gotta build it yourself lol | 14:59 |
| JayF | actually, you did | 15:00 |
| TheJulia | well, yeah :) | 15:00 |
| JayF | #startmeeting ironic | 15:00 |
| opendevmeet | Meeting started Mon Jan 30 15:00:20 2023 UTC and is due to finish in 60 minutes. The chair is JayF. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
| opendevmeet | The meeting name has been set to 'ironic' | 15:00 |
| JayF | Good morning, who all is here? | 15:00 |
| iurygregory | o/ | 15:00 |
| dtantsur | o/ | 15:00 |
| TheJulia | o/ | 15:00 |
| rpittau | o/ | 15:01 |
| * TheJulia stumbles across the channel saying "brraaaaainns" | 15:01 | |
| JayF | #topic Announcements | 15:01 |
| JayF | As always, please hashtag your patches #ironic-week-prio when they are ready for review, and use the priority dashboard in the whiteboard to do reviews. | 15:01 |
| JayF | #link https://ironicbaremetal.org/blog/baremetal-sig-2023q1/ | 15:01 |
| JayF | Q1 2023 Baremetal SIG is happening Weds, f | 15:02 |
| JayF | **Weds, Feb 8 from 1500-1700 UTC. | 15:02 |
| JayF | Please see the linked blogpost for details. | 15:02 |
| * TheJulia sighs since she will be in Dallas, TX | 15:02 | |
| JayF | Aight, next up | 15:02 |
| JayF | #topic review previous action items | 15:03 |
| JayF | I am going to push my action about LP; releasenotes job stole that time | 15:03 |
| JayF | #action JayF to bring a launchpad migration plan to a future meeting | 15:03 |
| JayF | I did fulfill the actions I had to 1) update bugfix branch policies (these have merged) and 2) clean up core list to remove now-inactive members :( | 15:04 |
| JayF | vanou had an item to update our security docs; I haven't seen that come across | 15:04 |
| JayF | vanou: do you have that or should we push it too? | 15:04 |
| kubajj | o/ | 15:04 |
| JayF | I don't think vanou is here; I'll push the action so we can check next week | 15:05 |
| TheJulia | I suspect vanou is asleep | 15:05 |
| JayF | #action vanou follow up on ironic security docs action from 1/23 meeting | 15:05 |
| JayF | That's it for action items. | 15:05 |
| JayF | #topic Review Ironic CI status | 15:05 |
| JayF | Ironic CI has been bonkers. | 15:05 |
| JayF | I fixed issues caused in the releasenotes job by the presence of the bugfix-x.y-eol tags -- we now configure reno to ignore them. | 15:06 |
| JayF | I know rpittau has continued to work on making some stable branches happy, too. Thanks for that. | 15:06 |
| JayF | I'd say I expect master should work, but some stable/bugfix branches may still be struggling | 15:06 |
| rpittau | JayF: I'm removing the reno jobs from the bugfix branches, doesn't make sense to have that there | 15:06 |
| JayF | ++ | 15:06 |
| opendevreview | Jakub Jelinek proposed openstack/ironic master: Erase swift inventory entry on node deletion https://review.opendev.org/c/openstack/ironic/+/871394 | 15:07 |
| JayF | if you find something wrong, please put it in the whiteboard or comment in IRC :) | 15:07 |
| JayF | Moving on | 15:07 |
| JayF | #topic review 2023.1 workstream progress | 15:07 |
| JayF | I'll note I haven't found much time for getting the last bit of sharding done; I'm going to try and hide somewhere today and complete it. | 15:07 |
| JayF | #link https://etherpad.opendev.org/p/IronicWorkstreams2023.1 | 15:08 |
| JayF | looks like sqlalchemy 2.0 is just waiting on the metal3 CI job | 15:08 |
| JayF | and merging inspector into ironic; kubajj has made great progress ... but I'm not sure it seems we'll get totally migrated this cycle? How is it looking? | 15:08 |
| dtantsur | Totally - absolutely not | 15:09 |
| dtantsur | the introspection data API is just the first step | 15:09 |
| TheJulia | Does a list or rough plan exist someplace? | 15:09 |
| dtantsur | yes, on storyboard. hold on | 15:09 |
| dtantsur | #link https://storyboard.openstack.org/#!/story/2010275 Plan on merging inspector in ironic | 15:10 |
| JayF | OK; that is super useful, thank you | 15:10 |
| TheJulia | I've added the link to the etherpad | 15:10 |
| JayF | TBH at this point I assume storyboard data is wrong unless someone tells me otherwise :| | 15:10 |
| JayF | Thank you for that work. We need to land much of this as the release will be here before we realise it :D | 15:11 |
| JayF | moving on | 15:11 |
| JayF | #topic Project Maintainer Audit @ PyPI | 15:11 |
| JayF | #link https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup | 15:11 |
| JayF | The results found for ironic in ^ can only be best described by an emoji screaming | 15:12 |
| JayF | We have multiple repos with maintainers whom are no longer associated with the community, and in one case, not identifyable at all | 15:12 |
| JayF | I am asking for unanimous consent to move forward with the TC plan to remove all existing maintainers except openstackci. | 15:12 |
| * iurygregory see his name in the list XD | 15:12 | |
| rpittau | absolutely, let's do it | 15:13 |
| JayF | TBH; I'm not sure this is something that we as a community haven't even already consented to, and it's the right thing to do, but I wanted to provide a venue in case there is any disagreement | 15:13 |
| TheJulia | ... *blink* *blink* | 15:14 |
| iurygregory | no objections from me | 15:14 |
| * TheJulia wonders how monty ended up a maintainer on bifrost | 15:14 | |
| JayF | generally speaking; it's whoever created the PyPI project originally | 15:15 |
| JayF | like I suspect the `login.launchpad.net_154` is probably one of my ex-managers at Rackspace, when we created it | 15:15 |
| JayF | (it being IPA) | 15:15 |
| JayF | I'm going to consider this agreed. | 15:16 |
| JayF | #agreed Ironic is onboard with TC plans to unify PyPA maintainership under openstackci | 15:17 |
| * dtantsur has no objections | 15:17 | |
| JayF | #undo | 15:17 |
| opendevmeet | Removing item from minutes: #agreed Ironic is onboard with TC plans to unify PyPA maintainership under openstackci | 15:17 |
| JayF | #agreed Ironic is onboard with TC plans to unify PyPI maintainership under openstackci | 15:17 |
| JayF | Aight, next up. | 15:18 |
| JayF | No RFEs to review, skipping that | 15:18 |
| JayF | #topic virtualpdu situation | 15:18 |
| JayF | rpittau: o/ | 15:18 |
| TheJulia | did they respond? | 15:18 |
| rpittau | yes, the status is not good | 15:18 |
| rpittau | they did not respond | 15:19 |
| JayF | So can we be clear how Ironic is impacted by the issue with VirtualPDU? | 15:19 |
| TheJulia | ++ | 15:19 |
| JayF | Just to reset back to the original trigger | 15:19 |
| TheJulia | I would like to understand that | 15:19 |
| rpittau | ironic jobs depend on virtualpdu | 15:19 |
| dtantsur | It's used in the SNMP job | 15:19 |
| TheJulia | yes | 15:19 |
| rpittau | virtualpdu is not python3 compatible | 15:19 |
| TheJulia | oh, heh | 15:20 |
| rpittau | we can't move to jammy | 15:20 |
| iurygregory | oh wow | 15:20 |
| rpittau | this is in short | 15:20 |
| TheJulia | ... and without an active maintainer.. ugh | 15:20 |
| JayF | So we really don't have a lot of time to resolve this either; then | 15:20 |
| rpittau | this https://review.opendev.org/c/openstack/ironic/+/869052 is failing because of that | 15:20 |
| dtantsur | which options we have? 1) fork virtualpdu, 2) put it in a container with old ubuntu, 3) drop the job? | 15:20 |
| JayF | 4) wait longer | 15:20 |
| TheJulia | 5) Ask to adopt it | 15:21 |
| TheJulia | ? | 15:21 |
| dtantsur | We cannot ask if they don't respond | 15:21 |
| JayF | TheJulia: #5 has been done; we've not gotten any response | 15:21 |
| dtantsur | If we decide to wait, the container approach can be a workaround | 15:21 |
| dtantsur | to proceed with jammy | 15:21 |
| rpittau | the project is abandoned since years now btw | 15:21 |
| JayF | Honestly, I think it makes sense to fork it | 15:21 |
| TheJulia | no response does sort of lend ourselves to lazy consensus ability | 15:21 |
| JayF | We're the only people in the world with a real need to reliably support fake-hardware emulation lol | 15:22 |
| dtantsur | Think about this: we can always remerge out fork if they wake up | 15:22 |
| rpittau | last update was 4 years ago https://opendev.org/x/virtualpdu | 15:22 |
| JayF | Yep. | 15:22 |
| JayF | So it's under opendev already? | 15:22 |
| rpittau | yeah | 15:22 |
| JayF | Does anyone have any idea what the process is for this? | 15:22 |
| JayF | and/or want to own the communication bits of this? | 15:22 |
| rpittau | I can handle that | 15:22 |
| TheJulia | there likely needs to be a dialog with the opendev folks to figure out what they are most comfortable with | 15:23 |
| rpittau | just need some tips :0 | 15:23 |
| JayF | I'd go into #opendev and tell them the situation | 15:23 |
| TheJulia | given being in the x/ namespace, it was once likely in openstack/ | 15:23 |
| TheJulia | ++ | 15:23 |
| JayF | that we've agreed in Ironic to adopt it | 15:23 |
| rpittau | sounds good | 15:23 |
| JayF | and we're willing to do that as a fork if needed | 15:23 |
| JayF | or not as a fork if hijacking is an option (I probably prefer not? but they might have context we don't) | 15:23 |
| JayF | #action rpittau to communicate with opendev/openstack folks to determine how best for Ironic community to adopt/fork https://opendev.org/x/virtualpdu | 15:24 |
| JayF | rpittau: goes without saying, if you have any trouble or confusion please ask :) I don't have much special knowledge but I can ask everyone :D | 15:25 |
| rpittau | of course :) | 15:25 |
| JayF | Anything else on virtualpdu? | 15:25 |
| rpittau | that's all I guess | 15:25 |
| JayF | #topic Open Discussion | 15:25 |
| JayF | anything for open discussion? | 15:25 |
| TheJulia | nothing from my point of view at this time | 15:26 |
| JayF | Aight, I'm going to call it. | 15:26 |
| JayF | #endmeeting | 15:26 |
| opendevmeet | Meeting ended Mon Jan 30 15:26:45 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:26 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-01-30-15.00.html | 15:26 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-01-30-15.00.txt | 15:26 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-01-30-15.00.log.html | 15:26 |
| TheJulia | iurygregory: o/ | 15:29 |
| iurygregory | o/ | 15:29 |
| TheJulia | got a little bit to chat about i-p-e? | 15:30 |
| iurygregory | going for lunch now, can be when I'm back? | 15:30 |
| TheJulia | sure | 15:30 |
| opendevreview | Julia Kreger proposed openstack/ironic-prometheus-exporter master: WIP: Support extraction of ironic internal metrics https://review.opendev.org/c/openstack/ironic-prometheus-exporter/+/869509 | 15:34 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent bugfix/8.3: Fix CI https://review.opendev.org/c/openstack/ironic-python-agent/+/871980 | 16:07 |
| iurygregory | TheJulia, I'm back o/ | 16:18 |
| TheJulia | iurygregory: cool | 16:22 |
| * TheJulia looks for one of the windows she had open | 16:22 | |
| TheJulia | iurygregory: https://meet.google.com/vbm-vqku-mhe | 16:24 |
| TheJulia | and https://paste.openstack.org/show/b2AeycAXKHCNoa5rMyUp/ | 16:24 |
| iurygregory | ok | 16:25 |
| opendevreview | Julia Kreger proposed openstack/ironic-prometheus-exporter master: WIP: Support extraction of ironic internal metrics https://review.opendev.org/c/openstack/ironic-prometheus-exporter/+/869509 | 16:57 |
| TheJulia | iurygregory: https://paste.openstack.org/show/bzclDetApkB5UdslepcX/ | 17:06 |
| TheJulia | with that patch above :) | 17:06 |
| iurygregory | TheJulia, ack | 17:07 |
| rpittau | good night! o/ | 17:11 |
| TheJulia | goodnight rpittau | 17:12 |
| JayF | So I'm implementing querying ports by shards | 21:53 |
| JayF | and I found what appears to be a bug in our API code | 21:53 |
| JayF | HOWEVER fixing it could throw 400 errors to existing clients | 21:53 |
| JayF | (those clients wouldn't be getting the results they wanted anyway) | 21:53 |
| JayF | it proactively raises here if we specify both a node_ident and portgroup_ident: https://github.com/openstack/ironic/blob/master/ironic/api/controllers/v1/port.py#L254 | 21:54 |
| JayF | but in the actual if statement, just below in L257-285, it's obvious that *address* is mutually exclusive, too | 21:54 |
| JayF | since I'm implementing shards, and that would be mutually exclusive, it's trivial to make this case error instead of returning incorrect results, but I'm unsure if that's the right path | 21:55 |
| JayF | TheJulia: ^ thoughts? | 21:55 |
| JayF | tl;dr: if you query ports with both a node||portgroup AND an address, the address gets completely ignored | 21:57 |
| JayF | whereas correct behavior would indicate returning a 400 as it's a bad query | 21:57 |
| TheJulia | so node_ident is largely going to be populated, for the most part, via subclassed resources of /v1/nodes/node/node-1/portgroups/ | 21:58 |
| TheJulia | getting the collection asking for specifics | 21:59 |
| TheJulia | hmmm | 21:59 |
| TheJulia | yeah, you can't filter at that level | 21:59 |
| TheJulia | so, I *think* it is valid to add shard to that list as well to just raise 400 | 21:59 |
| JayF | so there /is no/ current way for someone to say | 21:59 |
| JayF | for instance | 21:59 |
| JayF | /v1/ports?node_ident=some_uuid&address=some-mac | 21:59 |
| TheJulia | so that is a different query path | 22:00 |
| TheJulia | hmmmm | 22:00 |
| JayF | that doesn't call _get_ports_collection? | 22:00 |
| TheJulia | looking | 22:00 |
| TheJulia | it would be /v1/ports/?node=node-1&address=blah | 22:01 |
| JayF | that still populates node_ident at that level afaict, yeah | 22:02 |
| TheJulia | address is still valid | 22:02 |
| TheJulia | it is portgroup_ident which is for the subclassed resource view | 22:03 |
| JayF | not implemented that way, no | 22:03 |
| JayF | so if you look | 22:03 |
| JayF | you get in PortsCollection.get_all() | 22:03 |
| JayF | when node is passed in | 22:03 |
| JayF | it converts that to a node_ident which is passed to _get_ports_collection() | 22:03 |
| JayF | which is where we're at | 22:03 |
| TheJulia | oh | 22:03 |
| TheJulia | i see | 22:03 |
| TheJulia | https://github.com/openstack/ironic/blob/4e8705dc9f6219b0af8d6d2769fb5aa554f458c5/ironic/api/controllers/v1/port.py#L257-L285 | 22:03 |
| TheJulia | parameter precedence | 22:03 |
| JayF | exactly | 22:04 |
| JayF | if you set node and address, address is ignored | 22:04 |
| TheJulia | correct | 22:04 |
| JayF | when an operator might assume, based on our api-ref, that querying by node+address would return a port ONLY if the port's address matched | 22:04 |
| JayF | which breaks our API contract | 22:04 |
| JayF | so I lean towards making it a 400 ... but maybe hiding that behavior behind a microversion :| | 22:05 |
| TheJulia | I think we would need to double check the precise wording of the docs | 22:05 |
| TheJulia | conundrum here is node_ident is used by >1 means | 22:05 |
| JayF | https://docs.openstack.org/api-ref/baremetal/?expanded=list-ports-detail#id83 | 22:05 |
| JayF | TheJulia: is your suggestion we might need to actually hook up support for address+node_ident? | 22:06 |
| TheJulia | no | 22:07 |
| TheJulia | I think we need to just explicitly say "this is how it works" | 22:07 |
| JayF | So you're saying we should retcon the docs and call it expected behavior, more or less? | 22:07 |
| TheJulia | yeah | 22:07 |
| JayF | lol | 22:07 |
| JayF | it's the obviously right thing to do | 22:07 |
| TheJulia | because unwinding that is going to be crazy | 22:08 |
| TheJulia | and breaking in multiple ways :( | 22:08 |
| JayF | if I couldn't fix it with the docs; I'd probably go the route of | 22:08 |
| TheJulia | since we would need to delineate everything out and if the caller is specific | 22:08 |
| TheJulia | and that would spaghettize some of the code path too | 22:08 |
| JayF | putting a check for if address+node_ident is set to bail | 22:08 |
| JayF | but I like fixing the docs a lot better lol | 22:09 |
| TheJulia | conundrum is that would break way down elsewhere | 22:09 |
| * TheJulia doublechecks something esle | 22:09 | |
| TheJulia | https://docs.openstack.org/api-ref/baremetal/?expanded=list-ports-detail,list-ports-by-node-detail#node-management-nodes | 22:10 |
| TheJulia | address is not listed there :) | 22:10 |
| TheJulia | nor here: https://docs.openstack.org/api-ref/baremetal/?expanded=list-ports-detail,list-ports-by-node-detail,list-portgroups-by-node-detail#node-management-nodes | 22:10 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Fix API docs to document port filtering behavior https://review.opendev.org/c/openstack/ironic/+/872229 | 22:13 |
| JayF | TheJulia: look under /v1/ports, not /v1/ports/detail | 22:13 |
| TheJulia | JayF: I know, but those other paths is that same code path | 22:14 |
| JayF | I'm not sure I understand why this is relevant? | 22:14 |
| JayF | address being None is OK is all cases | 22:14 |
| JayF | address being not-none and node/node_uuid/portgroup being set is not great | 22:15 |
| TheJulia | it is the code path why it is dual used code | 22:15 |
| JayF | and what I updated the docs to improve | 22:15 |
| TheJulia | because of parent node_idents or portgroup_ident | 22:15 |
| JayF | yeah, that's why node_ident can be passed in at api query or class creation time, yeah | 22:15 |
| JayF | either way, I think my patch ^^ covers it | 22:16 |
| JayF | I'm going to take feedback on that patch and move back to making sharding work | 22:16 |
| TheJulia | ++ | 22:16 |
| vanou | good morning ironic | 23:14 |
| vanou | JayF: I'm working on writing security doc. I think I can propose draft this week. Sorry for slow progress. | 23:15 |
| JayF | vanou: it's not urgent; we were just following up on the action item in the meeting | 23:15 |
| JayF | vanou: since you're asleep then, if you wanna put an update for it in the meeting agenda (under follow up on action items) you can; or I'll just keep a lookout for this :) | 23:16 |
| JayF | (do not wake up early / stay up late to go to the meeting though; sleep is important :D) | 23:16 |
| vanou | Understand. Thanks a lot :) | 23:18 |
| JayF | Where are you in the world, vanou | 23:19 |
| JayF | just curious | 23:19 |
| JayF | I'm in Washington state in the USA, top left corner of the US. :) | 23:19 |
| vanou | I'm in Japan | 23:19 |
| JayF | That's what I assumed but wasn't sure :D | 23:20 |
| * TheJulia misses Japan | 23:20 | |
| vanou | DC is place I want to go. Good place. | 23:20 |
| vanou | JayF TheJUlia Have you gone Japan? | 23:21 |
| JayF | I had a chance to go many years ago when the summit was in Tokyo; I passed | 23:21 |
| JayF | I am a large man even by US standards (6' over 300 lbs), so I didn't think Tokyo would be well suited for me :D | 23:21 |
| TheJulia | I've been to Japan twice, First time was for the Summit, the second time was for a vacation. I spent lots of time roaming around tokyo by foot | 23:22 |
| * TheJulia goes and jumps in the car for a few hours of driving :( | 23:23 | |
| JayF | I have been around the world a bit; Japan would be a place I'd like to visit but that's probably 100 lbs and many many years away :D | 23:23 |
| vanou | JayF TheJulia: Nice :) Maybe summit related to OSS? | 23:24 |
| JayF | We'll see :D I can really only travel freely during the summer; going to conferences and such while school is in session is difficult for us right now | 23:24 |
| JayF | (my wife is a teacher and we have animals that can't be left alone/with a sitter) | 23:25 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Allow port queries by shard list https://review.opendev.org/c/openstack/ironic/+/872235 | 23:26 |
| JayF | don't get excited, ^ it's a wip that needs tests :P | 23:26 |
| vanou | JayF: I see. Animal family gives nice life. I have cat too. He is important part of my life. | 23:26 |
| JayF | but everything is hooked up end to end | 23:26 |
| JayF | vanou: yeah, we have a cat who has cancer and gets medicine daily, and a dog who can't be left alone all day :/ | 23:27 |
| JayF | (The sick cat is 100% in remission right now, and is doing incredibly well; but he still gets the medicine) | 23:27 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Allow port queries by shard list https://review.opendev.org/c/openstack/ironic/+/872235 | 23:28 |
| vanou | JayF: Then long/short travel should be avoided, naturally. I wish his health better | 23:29 |
| vanou | TheJulia: Regarding backport patch https://review.opendev.org/c/openstack/ironic/+/870880. This patch consists of 2 part: (1)Verify step to check HTTP(S) connection possible (2)Vendor interface which allows fetching iRMC firmware version. To deal with firmware HTTP(S) incompatibility which commit message tells, only first part of code is required. Second part of code was to | 23:36 |
| vanou | be | 23:36 |
| vanou | used to deal with firmware IPMI incompatibility storyboard tells. I think adding new vendor interface in backport patch is not good, as you said. So I think I remove second part of code from | 23:36 |
| vanou | backport patch. How do you t | 23:36 |
| vanou | ^ How do you think about it? | 23:37 |
| JayF | You'd still keep the logic though, yeah? So existing operators would be seeing new traffic to their nodes (e.g. using IPMI in some cases that it's not currently?) ... or am I confusing multiple patches? | 23:45 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!