15:01:42 <adamw> #startmeeting Fedora QA meeting
15:01:42 <zodbot> Meeting started Mon Aug 29 15:01:42 2022 UTC.
15:01:42 <zodbot> This meeting is logged and archived in a public location.
15:01:42 <zodbot> The chair is adamw. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
15:01:42 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:01:42 <zodbot> The meeting name has been set to 'fedora_qa_meeting'
15:01:46 <adamw> #meetingname fedora-qa
15:01:46 <zodbot> The meeting name has been set to 'fedora-qa'
15:01:53 <adamw> #topic Roll Call
15:01:56 <adamw> morning folks
15:02:00 <bittin-> afternoon
15:02:04 * SumantroMukherje is here
15:02:11 * bittin- is also here
15:02:16 <SumantroMukherje> morning adamw!
15:02:35 <tflink[m]> morning
15:03:16 * coremodule is here
15:03:21 <coremodule> good morning!
15:03:49 <adamw> how's everyone doing?
15:04:04 * bittin- is alright
15:05:54 <adamw> good good
15:06:02 <adamw> #topic Previous meeting follow-up
15:06:32 <adamw> "coremodule to remove 32-bit arm columns from validation matrix templates" - how's that going?
15:06:41 <coremodule> done!
15:07:21 <adamw> yay
15:07:31 <adamw> my favorite kind of follow-up
15:07:40 <adamw> #info "coremodule to remove 32-bit arm columns from validation matrix templates" - this is done
15:07:49 <adamw> if you see any stray 32-bit arm stuff in validation test pages, yell
15:07:54 <adamw> any other follow-up anyone?
15:10:12 <adamw> alrighty then
15:10:36 <adamw> oh hmm
15:10:40 <adamw> oh yes tim's here good
15:10:43 <adamw> #topic Kiwi TCMS evaluation results
15:10:46 <adamw> #chair tflink
15:10:46 <zodbot> Current chairs: adamw tflink
15:10:49 <adamw> take it away, tflink
15:11:42 <tflink[m]> I do recognize some irony here but do we want to do this topic this week? I know kparal is out and AFAIK, he's the only other person who tried out the Kiwi demo instance
15:11:46 <geraldosimiao> .hello geraldosimiao
15:11:47 <zodbot> geraldosimiao: geraldosimiao 'Geraldo S. Simião Kutz' <geraldo.simiao.kutz@gmail.com>
15:12:34 <adamw> tflink: i'm easy...
15:13:14 <coremodule> tflink[m], I *did* help with your setup tflink[m], not that I have much input here, but take that for what it's worth
15:14:14 <tflink[m]> I don't think that the outcome of the conversation is going to be different next week, anyways
15:14:20 <tflink[m]> #link https://lists.fedoraproject.org/archives/list/test@lists.fedoraproject.org/message/HSNAKQHR2TBQFYJAEMD4SWATJFA4WHT6/
15:14:37 <cmurf> oops i'm actually here
15:14:56 <tflink[m]> so, there has been some discussion around test case management and potentially moving away from using the wiki for test cases and results
15:15:21 <adamw> cmurf: hi
15:15:40 <cmurf> adamw hi
15:15:48 <bittin-> is that also for test days if that happends? *curious*
15:16:13 <tflink[m]> as KiwiTCMS is the only maintained open source option for test case management, I set up a demo instance to see if kiwi was a good choice for Fedora QA
15:16:41 <tflink[m]> bittin-: yeah, that would change as well
15:16:48 <tflink[m]> or at least that was part of the idea
15:16:55 <bittin-> ah thanks, i will let you go on
15:18:18 <tflink[m]> kparal set a long summary to test@ about his thoughts after poking around with the demo and it seems like his thoughts on the matter mostly match mine
15:19:04 <tflink[m]> which is: There's nothing horribly wrong with Kiwi as a TCMS but its implementation details make it a not very good choice for Fedora
15:19:45 <tflink[m]> we could switch over to using it but the cost of porting everything, changing all of our documentation and teaching all the new ways of doing things would be very expensive
15:20:50 <adamw> yeah, that was my take on yours and kparal's takes
15:20:53 <tflink[m]> so, if we don't switch to Kiwi, that leaves us with two options that I'm aware of: keep using the wiki or write something from scratch
15:21:03 <adamw> sorry i never got around to playing with it myself, but it sounds a whole lot like my experience with moztrap
15:21:53 <tflink[m]> both have good and bad things - as much of an odd use case as it is, the wiki works for us for the most part and doesn't require much in the way of upkeep
15:22:09 <bittin-> same haven't had time to test it myself, but i will use whatever the Fedora community decides to use
15:22:56 <tflink[m]> writing something from scratch would certainly cover all the things we do that are less common and integrate nicely with existing Fedora systems but it'd take time to create and maintenance once it's created
15:23:09 <tflink[m]> would that time be better spent doing something else?
15:23:45 <tflink[m]> that's the primary question that I wanted to ask here - where do we go or at least look at going from here?
15:25:23 <adamw> so we have several newer contributors here
15:25:42 <adamw> how painful do you folks find it to report results in the wiki? is it a real pain or just a little thing you got used to?
15:25:48 <tflink[m]> The primary concern raised in the test@ thread was about the usability of the test matrices and how easy it is to break them if you're not used to them already
15:26:37 <bittin-> the Wiki works for me, i am not the best with Wiki syntax however, but can always copy the person before only done so like 2-3 times, mostly used the testweeks. app
15:28:01 <adamw> oh, yeah, we do also have the option of kinda extending the testdays webapp to cover validation too. dunno how sane that'd be.
15:28:29 <adamw> off the top of my head, its requirement to create metadata pages would be a stumbling block. but we could...i dunno. mmm.
15:29:06 <SumantroMukherje> adamw, multi arch same test case on metapage is weird :)
15:29:16 <adamw> yeah
15:29:25 <adamw> we'd have to do quite a bit of extending...
15:29:36 <tflink[m]> at that point, I'd almost rather have a dedicated frontend - using wiki pages as a database seems like a very odd way of doing things, especially if folks can edit the wiki page and whatever frontend we're using
15:29:50 <tflink[m]> dedicated frontend and database backend, rather
15:30:22 <tflink[m]> unless tweaking testdays and/or the wiki layout would be easier and less problematic than I think it would be
15:31:47 <geraldosimiao> <adamw> "how painful do you folks find it..." <- I use relval to fill the release validation wiki. Simple and straightforward.
15:32:38 <adamw> yay, i like good reviews
15:33:02 <tflink[m]> another interesting (but potentially problematic) question is that if the current wiki-based setup is intimidating new users, would we ever hear about it?
15:33:06 <bittin-> geraldosimiao: need to look into that
15:33:20 <tflink[m]> problematic in that it promotes wild speculation
15:33:38 <adamw> tflink: i'm hoping we'd hear from folks joining up then telling us 'oh god this wiki thing is terrible' before they leave. but of course, hard to prove.
15:34:19 <adamw> bittin-: relval is a dumb little CLI client for the wiki stuff that i wrote. `relval report-results` walks you through reporting validation results and does the wiki editing for you so you can't mess it up.
15:34:55 <bittin-> adamw: geraldosimiao: cool did not know about that, need to look at that next time thanks for the tip/info :)
15:35:00 <SumantroMukherje> tbh, I like wiki and how easy it is to onboard people with relval . Thats my take on wiki at least.
15:35:27 <adamw> in the end it boils down to 'is writing a new thing from scratch the best use we can think of for the dev hours', i guess
15:35:28 <bittin-> wrote a reminder for myself
15:35:45 <tflink[m]> adamw: pretty much, yeah
15:35:53 <adamw> for myself i can think of at least five things i'd put a higher value on the time for, but i dunno about anyone else
15:36:02 <tflink[m]> and it sounds like the consensus here is pretty much: probably not
15:36:16 <TheExorcist[m]> .hi
15:36:17 <zodbot> TheExorcist[m]: Sorry, but user 'TheExorcist [m]' does not exist
15:36:34 <TheExorcist[m]> .hi naraiank
15:36:35 <zodbot> TheExorcist[m]: Sorry, but user 'TheExorcist [m]' does not exist
15:36:46 <tflink[m]> the wiki has it's issues and quirks but it does pretty much everything we need it to do and any replacement would also have issues and quirks
15:36:48 <adamw> it's ok, we know who you are :D
15:37:01 <TheExorcist[m]> .hello
15:37:01 <zodbot> TheExorcist[m]: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1".
15:37:08 <tflink[m]> .hello2 tflink
15:37:09 <zodbot> tflink[m]: Sorry, but user 'tflink [m]' does not exist
15:37:17 <tflink[m]> that's probably what you're looking for
15:37:35 <tflink[m]> never mind
15:37:46 <adamw> i think you need .hello if you need to put in the username manually
15:37:54 <adamw> .hello2 uses your nick
15:38:05 <SumantroMukherje> .hello sumantrom
15:38:05 <adamw> ...or something
15:38:05 <tflink[m]> I had it backwards, whoops
15:38:06 <zodbot> SumantroMukherje: sumantrom 'Sumantro Mukherjee' <sumukher@redhat.com>
15:38:54 <tflink[m]> anyhow, are there any dissenting opinions on replacing the wiki as a TCMS?
15:39:45 <coremodule> the wiki works for me at this time.
15:39:58 <tflink[m]> proposed #agreed the wiki may not be perfect but it is likely better than the alternatives of moving everything over to kiwi or writing our own TCMS from scratch
15:40:23 <TheExorcist[m]> tflink[m]: May be it needs to be forwarded to Council, I guess. Correct me, if I'm wrong.
15:41:00 <tflink[m]> The Exorcist: I suspect that the council would defer to our judgement here
15:41:20 <bcotton> Yeah, if it went to the Council, my first question would be "why are you asking us?"
15:41:41 <adamw> yeah, this is more of an our team thing
15:43:02 <Penguinpee> off topic: Is it intentional that kiwi is an anagram for wiki?
15:43:32 <tflink[m]> I suspect so
15:43:46 <tflink[m]> sorry, I misread
15:43:53 <tflink[m]> I suspect it's a coincidence
15:45:34 <tflink[m]> OK, it sounds like the people here are pretty much in agreement
15:45:59 <tflink[m]> #agreed the wiki may not be perfect but it is likely better than the alternatives of moving everything over to kiwi or writing our own TCMS from scratch
15:46:03 <adamw> yup
15:46:10 <adamw> thanks for the work on the testing, tflink
15:46:15 <tflink[m]> I'll send out a conclusion to the thread on test@
15:46:21 <bittin-> :) +1
15:46:28 <bittin-> +1 tflink[m]
15:46:34 <bittin-> .cookie tflink[m]
15:46:37 <bittin-> .cookie tflink
15:47:00 <SumantroMukherje> tflink++
15:47:06 <bittin-> tflink++
15:47:12 <bittin-> ah i forgot how you did it
15:47:27 <bittin-> so on to some status then?
15:48:20 <adamw> #topic Fedora 37 check-in
15:48:28 <adamw> so, afaik, status of f37 is 'mostly working'
15:49:06 <bittin-> Workstation works in my testing VM in Virtualbox and new composes are rolling, but there might be some bugs but i guess we are talking about those in the Blocker bugs meeting later today
15:49:22 <adamw> coremodule: oh hmm, the Base validation page still has a "Release-blocking environments (armhfp)" section?
15:49:36 <TheExorcist[m]> were there any updates released for F37 PR testing? Did I messed them?
15:49:38 <adamw> oh, you changed it since we last created an event
15:49:52 <TheExorcist[m]> s/messed/missed/
15:50:13 <bittin-> TheExorcist[m]: https://openqa.fedoraproject.org/nightlies.html
15:50:35 <bittin-> was two sets of updates just today
15:51:13 <bittin-> seems some IoT 37 composes is broken otherwise rest seem to work
15:52:00 <adamw> we fixed that today
15:52:06 <bittin-> great :)
15:52:11 <adamw> there's still an outstanding bug with greenboot though
15:52:22 <bittin-> oh ok
15:53:20 <bittin-> #info bugs in greenboot for Fedora IoT 37
15:53:29 <geraldosimiao> adamw: I must redo download and iso check for the workstation build because 20220826 build didn't worked at pendrive usb live session
15:53:49 <geraldosimiao> Strange because same uso worked at vm
15:53:56 <adamw> geraldosimiao: if previous ones were ok, it's probably a glitch, off the top of my head i don't think anything relevant should have changed...
15:54:01 <geraldosimiao> s/uso/iso/
15:54:42 <geraldosimiao> adamw: Yeah, must be
15:54:42 <TheExorcist[m]> bittin-: Keep the updates posted in #quality:fedoraproject.org
15:54:55 <adamw> now we're in freeze it would be good if folks can run the basic and beta validation tests and find any bugs now, so we have plenty of time to fix them
15:55:08 <bittin-> TheExorcist[m]: don't have the time there is new images 1-3 times per day
15:55:14 <adamw> looks like the usual suspects that haven't been done yet - cloud testing on real clouds, the printing test case, stuff like that
15:55:29 <bittin-> just sign up to the testing-reports and testing mailinglists TheExorcist[m]
15:55:46 <adamw> #info it'd be great if folks can run the missing Basic and Beta validation tests for the current candidate compose so we can catch any blocker bugs with plenty of time to fix them
15:55:54 <adamw> i'll also send out an email about that
15:56:00 <SumantroMukherje> I would like to schedule a Beta Validation Week .. when do we start?
15:56:19 <bittin-> SumantroMukherje: the week after Beta drops maybe?
15:56:45 <adamw> Sumantro Mukherjee: do you mean pre-Beta testing, or post-Beta testing?
15:56:54 <adamw> if you mean pre-Beta testing, next week would be great :D
15:57:02 <SumantroMukherje> pre-beta, i suppose!
15:57:12 <SumantroMukherje> post beta will be pre-final :D
15:57:23 <bittin-> next week is also the i18n and GNOME 43 RC test weeks
15:57:41 <SumantroMukherje> so next week, we will be running "validation events" for the first time :D
15:57:54 <bittin-> a lot of testing next week
15:58:02 <adamw> hmm
15:58:14 <adamw> then this week would've been great, but i guess it's short notice :D
15:58:20 <adamw> the target release date is 13th sept
15:58:26 <adamw> so we can't really leave it later than next week
15:58:38 <adamw> let's switch topics while we still have two minutes
15:58:46 <adamw> #topic Test Day / community event status
15:58:48 <SumantroMukherje> I18n Test Week is coming up on 5th Sept through the end of the week.https://testdays.fedoraproject.org/events/139.  CryptoPolicy test day is also going to be on 5th Sept. IoT and Gnome Final to decide the dates. I will push in Gnome Apps, toolbx , virt, FCOS and podman somewhere in next come of weeks, before Fedora 37 upgrade test day.
15:58:50 <adamw> any really quick news/announcements, sumantro?>
15:59:01 <adamw> haha, you're ahead of me
15:59:15 <bittin-> GNOME 43 RC Test week starts 7th September and goes on until 14th September
15:59:17 <adamw> #info i18n Test Week is coming up on 5th Sept through the end of the week: https://testdays.fedoraproject.org/events/139
15:59:26 <adamw> #info CryptoPolicy test day is also going to be on 5th Sept.
15:59:43 <adamw> #info others are pending date decisions
16:00:25 <SumantroMukherje> so adamw, should we start the beta validation from 31st itself?
16:00:48 <adamw> sure, that'd be great if folks are readu
16:00:51 <adamw> ready*
16:00:57 <adamw> starting up the blocker bug review meeting now
16:01:11 <bittin-> works for me
16:01:17 <SumantroMukherje> I will write some doc and set things up!
16:02:38 <bittin-> f37 blocker review meeting starting now over in #blocker-review:fedoraproject.org
16:03:14 <adamw> awesome, thanks sumantro
16:03:27 <adamw> #info late notice Beta validation test week starting on August 31st
16:03:27 <SumantroMukherje> np adamw!
16:03:28 <adamw> #topic Open floor
16:03:34 <adamw> anyone got anything else burningly urgent?
16:03:36 * bittin- have nothing
16:05:12 <adamw> alrighty, thanks for coming everyone
16:05:31 * bittin- heads to the blocker review meeting
16:05:44 <adamw> #endmeeting