fedora-server
LOGS
18:00:14 <pboy> #startmeeting fedora-server
18:00:14 <zodbot> Meeting started Wed Jan 31 18:00:14 2024 UTC.
18:00:14 <zodbot> This meeting is logged and archived in a public location.
18:00:14 <zodbot> The chair is pboy. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
18:00:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:00:14 <zodbot> The meeting name has been set to 'fedora-server'
18:00:26 <pboy> #topic Welcome / roll call
18:00:38 <pboy> Welcome to our Server WG IRC meeting today!
18:00:38 <pboy> This is an out of order meeting to follow up on our Jan 19 meeting.
18:00:38 <pboy> „Same procedure as every year“ We'll give a few minutes for folks to show up
18:03:40 <jwhimpel> .hello2
18:03:41 <zodbot> jwhimpel: jwhimpel 'John Himpel' <john@jlhimpel.net>
18:04:22 <pboy> Hi jwhimpel !
18:05:42 <jwhimpel> Good morning from St. Louis, MO USA.  Nice sunny day here (after 8 days of overcast clouds).
18:07:06 <cooltshirtguy> .hello2
18:07:07 <zodbot> cooltshirtguy: cooltshirtguy 'Jason Beard' <jas_beard@hotmail.com>
18:07:12 <pboy> Well, it's good evening here, in Northern Germany. Luckily without rain,
18:07:30 <pboy> Hi cooltshirtguy!
18:07:49 <cooltshirtguy> Hello! Sunny and 70F+ down here in texas
18:08:52 <pboy> I see, quite warm, relative to Northern Germany
18:09:35 <cooltshirtguy> I prefer the warmth. :)
18:10:30 <cooltshirtguy> I saw in the news that the German Railway is looking for DOS/Windows 3.11 admin.
18:11:22 <pboy> That's a joke, isn't it?  Although, not impossible :-)
18:12:00 <pboy> Well, we have a small line-up today.
18:12:09 <pboy> Nevertheless, let's get started and see how far we get
18:12:21 <pboy> #topic Agenda
18:12:23 <cooltshirtguy> It's been in the news. :)
18:12:42 <pboy> #info  LVM2 default configuration change
18:12:42 <pboy> #info  Test planning for F40
18:12:42 <pboy> #info   A "story" for each release)
18:12:42 <pboy> #info  Work program and goals for F40
18:12:42 <pboy> #info  Revisiting Fedora Server release criteria
18:12:42 <pboy> #info  Open Floor
18:13:21 <pboy> cooltshirtguy  20 years ago?
18:13:39 <pboy> Any topic to add?
18:13:56 <jwhimpel> no additions here
18:14:03 <pboy> OK
18:14:13 <cooltshirtguy> nope
18:14:13 <pboy> #topic 1. LVM2 default configuration change
18:14:22 <pboy> was  Aarch64 SBC Installation bug on Fedora Server
18:14:32 <pboy> #link https://bugzilla.redhat.com/show_bug.cgi?id=2258764
18:14:32 <pboy> #link https://bugzilla.redhat.com/show_bug.cgi?id=2246871
18:14:32 <pboy> #link https://bugzilla.redhat.com/show_bug.cgi?id=2247872
18:14:32 <pboy> #link https://lists.fedoraproject.org/archives/list/server@lists.fedoraproject.org/thread/EEZAHSE5QSYSFVX5D5WLEJRJJ77YMVWK/
18:14:54 <pboy> Unfortunately, we are messing eseyman today
18:15:29 <pboy> Taking a short break to read?
18:15:59 <jwhimpel> What would be involved in a "fix"?  Just a PR to a script or is the issue deeper than that?
18:17:09 <pboy> Currently, the proposal is a modification in Anaconda. It will not create the /etc/lvm/devices/device.list file
18:17:25 <pboy> see the 3rd bugzilla entry
18:18:20 <pboy> I think that removes the symptom, but not the cause. We will be able to install on SBCs.
18:18:41 <pboy> And we might be able to use vgscan as used to.
18:19:21 <pboy> But another member of the LVM group did advice against removing that file, because we would miss some LVM properties.
18:19:35 <pboy> But they didn't say which ones.
18:20:41 <pboy> And currently, we don't know whether any operation  recreates the file later.
18:21:02 <cooltshirtguy> wow what a mess for one setting
18:21:32 <pboy> Yeah, according to my experience, there are operations that re-create that file.
18:22:00 <pboy> And then standard LVM administrative tasks would be impossible
18:22:15 <pboy> not all, but some
18:23:03 <pboy> unfortunately, noone of the LVM maintainers gave more information so far
18:24:14 <pboy> As fas as I know, upstream did create that device file, but set the config to ignore it (at least for vgscan and vgchange)
18:24:51 <jwhimpel> The LVM folks may not be following the ticket as the component is anaconda.
18:25:00 <pboy> @nirik are you available?
18:25:30 <nirik> in the middle of some datacenter work... can try and answer, whats up?
18:25:54 <pboy> Our discussion is about the LVM fix and what to do.
18:26:20 <pboy> Maybe, we should discuss it on mailing list, so we don't disturb you now?
18:26:21 <nirik> I've not followed that too closely. ;(
18:26:35 <nirik> but yeah, I was hoping there would be some fix before now...
18:27:22 <pboy> There is a PR to modify Anaconda, but that doesn't necessarly the cause and may have side effects.
18:27:48 <nirik> ah yeah... we need to try and get the lvm maintainers attention I guess?
18:28:09 <pboy> Yes, we tried, without success for know
18:28:19 <nirik> I can add it to my todo list, but it's already really long. ;)
18:28:48 <pboy> i think, we have a bit ot time, Beta is in 4 weeks.
18:29:18 <nirik> sure, but it's been this way since before last release... so really we need something. if only the anaconda workaround
18:29:31 <pboy> Proposal, we open a thread on mailing list and nirik adds it to his todo list.
18:29:58 <pboy> So we may have an informed proposal in aboutg 2 weeks?
18:29:59 <nirik> sure, I have already added it. Not sure when I will get there, but will try.
18:30:37 <jwhimpel> there is a lvm repository with 32 open issues.  It is at https://github.com/lvmteam/lvm2
18:30:48 <pboy> OK. proposal #agreed  We create a thread on mailing list to discuss the option for LVM fix.
18:31:22 <pboy> jwhimpel That doesn't look like a quick solution .-)
18:31:51 <pboy> any objections agsinst the proposal?
18:32:09 <pboy> 3
18:32:13 <pboy> 2
18:32:17 <pboy> 1
18:32:38 <pboy> #agreed  We create a thread on mailing list to discuss the option for LVM fix.
18:33:00 <pboy> #actionb pboy will create a thread on mailing list to discuss the option for LVM fix.
18:33:13 <pboy> #action  pboy will create a thread on mailing list to discuss the option for LVM fix.
18:33:25 <jwhimpel> The README there shows a mailing list at linux-lvm@lists.linux.dev  I'll try to raise the issue there also.
18:33:44 <pboy> jwhimpel  1++
18:34:26 <pboy> #action jwhimpel will try to contact linux-lvm@lists.linux.dev
18:34:52 <pboy> Anything else here?
18:35:09 <pboy> 3
18:35:14 <pboy> 2
18:35:18 <pboy> 1
18:35:30 <pboy> #topic 2. Test planning for F40
18:35:39 <pboy> #link https://pagure.io/fedora-server/issue/125
18:35:39 <pboy> #link https://lists.fedoraproject.org/archives/list/server@lists.fedoraproject.org/thread/WQVZ3JDOSUC3EBGGQIOYP4Y4FFS3CAPV/
18:36:25 <pboy> #link https://lists.fedoraproject.org/archives/list/server@lists.fedoraproject.org/thread/EEZAHSE5QSYSFVX5D5WLEJRJJ77YMVWK/
18:36:53 <jwhimpel> I was going to purchase an RPi 5 and test new install and dist upgrade.  However, Peter Robinson thinks RPi 5 is not yet ready for prime time.
18:37:25 <pboy> jwhimpel yes, we can do a lot of test on an SBC.
18:37:44 <pboy> On application level it is the same
18:38:29 <pboy> Hm, Fedora lists server seems to has some issues. I get Error 503 Backend fetch failed
18:38:45 <pboy> The last link is to my mailing list proposal today.
18:38:46 <cooltshirtguy> I got the same error on both
18:38:56 <jwhimpel> I can try an upgrade from F39 to F40 along with a F40 new install (NO LVM) using KVM.
18:40:11 <cooltshirtguy> are we testing on x86 hardware too?
18:40:19 <pboy> What a pity with the list server right now  :-)
18:41:00 <pboy> jwhimpel  upgrade test is very important. We have neglected this so far.
18:41:22 <pboy> How do we organise our efforts?
18:42:03 <pboy> Any idea?
18:42:51 <jwhimpel> I'll do a basic server install of F40 and a upgrade of a basic server upgrade from F39 to F40 on x86 using KVM (without LVM).
18:43:17 <pboy> jwhimpel ++1
18:43:48 <pboy> And can you try it with our VM image as well? So we would have LVM, too.
18:44:03 <pboy> Maybe, we'll have yet another surprise with LVM
18:44:54 <jwhimpel> Depends upon time available.  I have an out-patient surgery scheduled in a few days.  It may put me out of commission for several days.
18:45:25 <cooltshirtguy> what are testing, upgrades, new installs, applications ?
18:45:45 <pboy> jwhimpel good luck for the surgery. (I just had such an adventure the last 2 weeks)
18:46:22 <pboy> cooltshirtguy what d sou mean with "what are"??
18:46:58 <cooltshirtguy> sorry, grammar error. what are we specifically testing?
18:47:05 <jwhimpel> I'll try to concentrate on upgrade and new install on x86 VM.  I'll leave applications for others.
18:48:15 <pboy> cooltshirtguy  e.g nfs: pick our documentation and try to install and configure an nfs server and check if you gfet it working.
18:48:30 <pboy> You could do that with a VM on your server.
18:48:41 <pboy> Same with httpd.
18:50:41 <cooltshirtguy> I could
18:50:58 <pboy> Great!
18:52:25 <cooltshirtguy> when do we start testing? beta? release candidate?
18:52:26 <pboy> Regarding organization, what do we think about creating a table on our wiki page, containing the test item, a short test description and a row who can take over the task, or who cares about the task?
18:53:05 <pboy> cooltshirtguy  We should really start with beta latest. But better with the current F40 rawhide version.
18:53:29 <cooltshirtguy> ok
18:55:12 <pboy> Well, I think, I'll create a short draft in our wiki area and we can discuss the usability next meeting. Any objections?
18:55:19 <cooltshirtguy> nope
18:55:57 <jwhimpel> sounds good to me
18:56:07 <pboy> #action  pboy will create a draft wiki page to help to organise and coordinate our release testing efforts.
18:56:23 <pboy> Next topic:
18:56:37 <pboy> #topic 3. A "story" for each release
18:56:49 <pboy> We have to decide what to chose for F40.
18:56:49 <pboy> And we have to keep in mind, that we are supposed to write a release note about it.
18:56:49 <pboy> And probably initiate some other discussion, e.g. a blog post or a pod cast in cooperation with Fedora marketing.
18:58:47 <pboy> Latest discussion:
18:58:51 <pboy> #link https://pagure.io/fedora-server/issue/101
18:59:09 <jwhimpel> I don't think there are any significant changes for F40, so topics for a "story" are hard to come by.
19:00:23 <pboy> Well,. for 39 we decided to Fedora Server on SBC, but failed, unfortunately, and were hindered by the installation bug
19:01:02 <pboy> We could simple continue that project or we could choose one of the alternatives, we discussed.
19:01:29 <pboy> "story" is not necessarly about a change in the release.
19:02:49 <jwhimpel> Good to know
19:03:33 <pboy> Oh, I see our time is up
19:03:53 <cooltshirtguy> that was fast
19:04:19 <pboy> I suppose, we should continue asynchronous  on mailing list.
19:04:38 <pboy> cooltshirtguy  yes, we had a good discussion.
19:04:49 <jwhimpel> bye
19:05:02 <pboy> And then time passes very quickly  :-9
19:05:14 <cooltshirtguy> later
19:05:27 <pboy> OK, I close out meeting now and start discussion on mailing list.
19:05:53 <pboy> bye bye until NEXT WEEK  (Febr 7)
19:06:25 <pboy> #endmeeting