16:30:58 <dustymabe> #startmeeting fedora_coreos_meeting 16:30:58 <zodbot> Meeting started Wed Mar 22 16:30:58 2023 UTC. 16:30:58 <zodbot> This meeting is logged and archived in a public location. 16:30:58 <zodbot> The chair is dustymabe. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 16:30:58 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:30:58 <zodbot> The meeting name has been set to 'fedora_coreos_meeting' 16:31:03 <dustymabe> #topic roll call 16:31:05 <dustymabe> .hi 16:31:06 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dusty@dustymabe.com> 16:31:23 <gursewak> .hi 16:31:24 <zodbot> gursewak: gursewak 'Gursewak Singh' <gurssing@redhat.com> 16:31:40 <jlebon> .hello2 16:31:41 <zodbot> jlebon: jlebon 'None' <jonathan@jlebon.com> 16:31:47 <aaradhak> .hi 16:31:48 <zodbot> aaradhak: aaradhak 'Aashish Radhakrishnan' <aaradhak@redhat.com> 16:31:58 <marmijo[m]> .hi marmijo 16:31:59 <zodbot> marmijo[m]: Sorry, but user 'marmijo [m]' does not exist 16:32:32 <ravanelli> .hi 16:32:33 <zodbot> ravanelli: ravanelli 'Renata Ravanelli' <renata.ravanelli@gmail.com> 16:32:37 <apiaseck> .hi 16:32:39 <zodbot> apiaseck: Sorry, but user 'apiaseck' does not exist 16:32:45 <apiaseck> .hello c4rt0 16:32:53 <zodbot> apiaseck: c4rt0 'Adam Piasecki' <c4rt0gr4ph3r@gmail.com> 16:32:56 <marmijo[m]> .hello marmijo 16:32:57 <zodbot> marmijo[m]: marmijo 'Michael Armijo' <marmijo@redhat.com> 16:34:11 <dustymabe> #chair gursewak jlebon aaradhak marmijo[m] ravanelli apiaseck 16:34:11 <zodbot> Current chairs: aaradhak apiaseck dustymabe gursewak jlebon marmijo[m] ravanelli 16:34:14 <dustymabe> nice turnout today! 16:34:46 <dustymabe> now I expect all of you to speak up during the meeting! don't let us old timers dominate the conversation too much! 16:35:09 <jlebon> :) 16:35:21 <dustymabe> ok let's get started 16:35:26 <dustymabe> #topic Action items from last meeting 16:35:35 <dustymabe> #info no action items from last meeting! 16:35:41 <dustymabe> 🎉 16:36:01 <dustymabe> for meeting topics today I'll start with an FYI one 16:36:07 <dustymabe> #topic old bootloader versions don't boot new aarch64 6.2+ kernels 16:36:11 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/1441 16:36:55 <dustymabe> #info we rolled back the f38 update to our next stream and are in the process of re-releasing f38 into next after an intermediate barrier release that we put out yesterday. More information in https://github.com/coreos/fedora-coreos-tracker/issues/1441 16:37:15 <dustymabe> this one took some coordination. Thanks to everyone involved! 16:37:29 <jlebon> nice work all! 16:37:44 <dustymabe> I think it has also re-emphasized (for me at least) the value of barrier releases 16:38:35 <dustymabe> anyway.. moving on to next issue unless anyone wants to discuss this further 16:39:05 <dustymabe> #topic xfsprogs 6.1.0 disallows creating filesystems smaller than 300 MB 16:39:11 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/1443 16:39:23 <dustymabe> jlebon, want to intro this one since bgilbert isn't here 16:39:42 <jlebon> dustymabe: sure, can try 16:40:50 <jlebon> a new xfsprogs entered f37 which now prevents the creation of XFS filesystems smaller than 300 MB. we have some upstream tests that broke from this that can be adjusted. but this can also break anyone currently creating such filesystems from Ignition. 16:41:46 <jlebon> the breaking will happen anyway in f38, unless upstream revises its position 16:42:30 <jlebon> dustymabe: which part did you want to discuss more specifically? 16:42:51 <dustymabe> sorry - got pulled away 16:42:57 <dustymabe> I think benjamin added the meeting label 16:43:17 <dustymabe> I think he wants to know what the group thinks about the path forward 16:43:24 <dustymabe> i.e. do we need a status post? 16:43:43 <dustymabe> the good news is that this would only affect systems that are being newly provisioned 16:43:59 <dustymabe> so shouldn't affect upgrades at least 16:44:11 <dustymabe> well - at least the "ignition creating FS" part 16:45:17 <jlebon> hmm, not sure it's worth a status post 16:45:32 <dustymabe> yeah, probably not 16:45:44 <jlebon> we've only noticed this because of our upstream tests where we do artificial things 16:46:26 <dustymabe> do we agree for now that we keep it pinned in f37 and switch on the switch to f38? 16:46:29 <jlebon> i wouldn't be surprised if some users do create such small filesystems, but it's seems very uncommon 16:47:00 <dustymabe> there may be tools that create small FS for things (think libguestfs), but that isn't really our problem 16:47:10 <jlebon> this is the classic "bundle breakage vs spread it" question :) 16:48:14 <jlebon> keeping it pinned in f37 seems reasonable to me. people may be paying more attention during the f38 rebase. 16:48:51 <marmijo[m]> I think keeping it pinned in f37 sounds like a good plan 16:49:24 <dustymabe> #proposed we'll keep this pinned in f37 for now and introduce the change with the f38 rebase in each stream. we won't put out a coreos-status post as we think the use case is relatively uncommon. 16:49:43 * dustymabe thinks we need a "known issues" field in our release notes 16:49:47 <danpb> dustymabe: libguestfs appliance doesn't use xfs AFAIK 16:50:03 <dustymabe> 👋 danpb :) 16:50:16 <jlebon> ack 16:50:54 <dustymabe> all in favor? all oposed? 16:51:40 <gursewak> In favour 16:51:43 <marmijo[m]> SGTM 16:51:55 <dustymabe> #agreed we'll keep this pinned in f37 for now and introduce the change with the f38 rebase in each stream. we won't put out a coreos-status post as we think the use case is relatively uncommon. 16:52:12 <dustymabe> #topic tracker: Fedora 38 Test Week on March 28 16:52:17 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/1440 16:52:22 <dustymabe> ravanelli: want to intro this one? 16:52:32 <ravanelli> sure 16:53:24 <ravanelli> We had a meeting yesterday to enumerate new test cases, that are now described in https://github.com/coreos/fedora-coreos-tracker/issues/1445 16:54:09 <ravanelli> We now need volunteers to run the other tests too 16:54:17 <ravanelli> Or even help with this list 16:55:26 <ravanelli> The Test Day is helping next week. I will send the announcement soon 16:55:39 <dustymabe> ravanelli: it would be good if we could get people to agree to come to the test day meeting on the 28th to help us coordinate and execute the test cases 16:55:48 <dustymabe> anybody around/available next Tuesday? 16:56:13 <jlebon> o/ 16:56:49 * dustymabe notes that the `next` release we're building today and shipping tomorrow is the one we'll test during test day/week 16:56:56 <ravanelli> dustymabe: +1 I will try to find more people to help 16:57:06 <gursewak> I'm available next Tuesday. 16:57:26 <dustymabe> gursewak++ 16:57:28 <zodbot> dustymabe: Karma for gursewak changed to 1 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 16:57:54 <jlebon> "Installing FCOS on Azure aarch64" -- should that be "Booting FCOS on Azure aarch64" ? 16:58:44 <dustymabe> jlebon: yeah.. that one is also something we said we would "track for f39" since we don't have any docs for it right now (the aarch64 image creation process is slightly different than x86_64) 16:59:01 <dustymabe> hopefully for f39 we'll have images uploaded to azure already so no need to upload yourself 16:59:37 <jlebon> ack +1 16:59:37 <dustymabe> though, we should probably document for each cloud how to use our images or how to upload them yourself :/ 17:00:23 <jlebon> for s390x, it seems we already have docs to boot on IBM Cloud 17:01:13 <dustymabe> jlebon: right, but we need the test case created (the table on the test day page) 17:01:25 <dustymabe> SumantroMukherje: was going to that 17:01:45 <jlebon> i can take a couple from SumantroMukherje since it looks like he's got quite a few :) 17:02:04 <dustymabe> #info please join us for the test day next Tuesday March 28th. More information in https://github.com/coreos/fedora-coreos-tracker/issues/1440 17:02:07 <dustymabe> jlebon++ 17:02:07 <zodbot> dustymabe: Karma for jlebon changed to 1 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:02:23 <jlebon> SumantroMukherje: when you're around, let's chat in #fedora-coreos 17:02:31 <dustymabe> #topic kdump.crash test failing on aarch64 with kernel 6.2+ meeting 17:02:48 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/1430 17:02:52 <ravanelli> I think Nikita will help with s390x too 17:03:03 <dustymabe> I added the meeting label to this one. 17:03:28 <dustymabe> basically when we ship the 6.2 kernel to our prod streams we'll be breaking kdump 17:03:50 <dustymabe> I personally think this isn't a blocker, but am interested in other people's opinions 17:04:45 <jlebon> agree this shouldn't be a blocker 17:05:54 <dustymabe> #proposed The 6.2 kernel will introduce a regression for kdump on aarch64. While bad, we don't think this should blocking shipping kernel 6.2. 17:05:59 <dustymabe> votes ^^ 17:06:15 <jlebon> s/blocking/block/ 17:06:16 <jlebon> ack 17:06:35 <apiaseck> breaking silence: 17:06:36 <apiaseck> ack 17:06:56 <marmijo[m]> ack 17:06:57 <gursewak> ack 17:07:17 <dustymabe> #agreed The 6.2 kernel will introduce a regression for kdump on aarch64. While bad, we don't think this should blocking shipping kernel 6.2. 17:07:22 <dustymabe> thanks for the votes :) 17:07:28 <apiaseck> :) 17:07:31 <dustymabe> #topic Platform Request: kubevirt 17:07:36 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/1126 17:07:51 <dustymabe> I added the meeting label to this one.. 17:08:02 <dustymabe> gursewak and I are working on adding kubevirt to FCOS 17:08:31 <dustymabe> there is one item in the list (https://github.com/coreos/fedora-coreos-tracker/issues/1126#issuecomment-1452628138) that needs some discussion 17:08:53 <dustymabe> Push Kubevirt container image to quay: Where to push? 17:10:20 <dustymabe> we have the quay.io/fedora namespace 17:10:29 <dustymabe> which is where we push quay.io/fedora/fedora-coreos 17:11:00 <dustymabe> but https://quay.io/organization/containerdisks also exists (looks like it's trying to be a centralized repo for OS disk images) - backed by https://github.com/kubevirt/containerdisks 17:11:18 <dustymabe> thoughts on where we should push this stuff? 17:11:47 <jlebon> i feel like we discussed this in the past, when discussing the other container images we push 17:12:20 <jlebon> hmm, is that kubevirt repo given any sugar somehow on the k8s side? e.g. is it more convenient when images are there? 17:12:39 <dustymabe> jlebon: yeah I think we discussed it then too 17:12:45 <dustymabe> jlebon: I don't know the answer to your second question 17:13:09 <dustymabe> for the first question: https://github.com/coreos/fedora-coreos-tracker/issues/1171 17:14:03 <jlebon> huh, rhcos is there: https://github.com/kubevirt/containerdisks/blob/main/artifacts/rhcos/rhcos.go 17:14:38 <jlebon> that repo looks like glue that pulls from official disk image locations and converts it into containers? 17:14:40 <dustymabe> jlebon: yep 🤷♂️ 17:14:53 <jlebon> in our case, we're creating the container image ourselves 17:16:07 * dustymabe will note it's also nice that we control the repo we'd be using to distribute (i.e. bgilbert and I had to revert a tag update the other day) 17:17:20 * dustymabe puts down the microphone 17:17:29 <jlebon> indeed, yeah 17:17:45 <jlebon> would be good to talk to someone connected to that 17:17:57 <jlebon> s/that/that github repo/ 17:18:15 <dustymabe> we could invite fabian (or someone else) to our meeting next week 17:19:07 <jlebon> that'd be nice. i feel like we need more information on how that repo fits into the kubevirt UX 17:19:20 <dustymabe> ok I'll see what I can do on that front 17:19:31 <dustymabe> #topic open floor 17:19:36 <dustymabe> anyone with anything for open floor? 17:19:52 * dustymabe notes that Flock this year is in Ireland in August 17:20:02 <dustymabe> and the Devconf.us CFP is closing in two weeks 17:20:36 <jlebon> nice 17:21:52 <mnguyen> is anyone planning on doing any talks? 17:22:23 <mnguyen> feels like yesterday I was with dustymabe at devconf.us 17:22:31 <dustymabe> mnguyen: I know right 17:22:57 <dustymabe> mnguyen: would be nice, but I haven't thought about it yet (just submitted talks for devconf.cz) 17:23:18 <dustymabe> we should at least do the lab/workshop (at every conference we can). It's a good way to get new people hands on. 17:24:28 <dustymabe> any other topics for open floor? 17:25:50 <dustymabe> #endmeeting