14:02:00 <pbrobinson> #startmeeting Fedora IoT Working Group Meeting
14:02:00 <zodbot> Meeting started Wed Jan 15 14:02:00 2020 UTC.
14:02:00 <zodbot> This meeting is logged and archived in a public location.
14:02:00 <zodbot> The chair is pbrobinson. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:02:00 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:02:00 <zodbot> The meeting name has been set to 'fedora_iot_working_group_meeting'
14:02:00 <pbrobinson> #chair pwhalen pbrobinson bcotton jsmith
14:02:00 <zodbot> Current chairs: bcotton jsmith pbrobinson pwhalen
14:02:01 <pbrobinson> #topic roll call
14:02:16 * jsmith is here, but multitasking with ${DAYJOB}
14:02:19 * pwhalen is here
14:02:27 * bcotton is here
14:02:35 <pwhalen> good morning
14:02:40 * puiterwijk is here
14:02:49 <pbrobinson> morning all
14:03:26 <pwhalen> good evening to you pbrobinson
14:03:45 * tdawson is lurking this morning
14:03:50 <pbrobinson> #topic 1) ==== Working Group process and admin  ====
14:04:21 <pbrobinson> I'm sure there's stuff we need to cover here, but not coming to mind. bcotton or anyone else have anything?
14:04:42 <bcotton> well since you asked
14:05:26 <bcotton> i'm curious to know how well having a Bugzilla component for IoT has worked out. are bugs getting filed there? has it helped consolidate the many venues where people yell at peter to fix things?
14:06:11 <pwhalen> I havent filed any iot specific bugs since it was introduced
14:06:22 <pbrobinson> there's a few, mostly tracking bits
14:07:03 <pbrobinson> but then we have a few different venues for bugs and I don't feel any have been any more or less effective than any of the others. Maybe people are just confused
14:08:00 <bcotton> i'd say "let's update the website" but... :-)
14:09:04 <pbrobinson> bcotton: I think we list some of it in the docs, we should likely update that, I've been meaning on going through and reviewing the docs and updating it
14:09:09 * pbrobinson adds it to the list
14:12:21 <pbrobinson> anything else?
14:12:30 <pbrobinson> #topic 2) ==== Fedora Rawhide AKA F-32 ====
14:12:49 <pbrobinson> we have regular composes and it mostly seems OK
14:12:54 <pwhalen> Looking pretty good with the latest
14:13:09 <pbrobinson> and I know puiterwijk isn't far from provisioning bits being usable
14:13:30 <puiterwijk> Yeah, I was writing the user docs until this meeting started
14:13:53 <pwhalen> nice!
14:13:54 <pbrobinson> cool
14:14:01 * pwhalen looks forward to testing it out
14:14:44 <puiterwijk> If people are willing to help test it off-the-grid and give me feedback on the docs while I'm writing them, I'll give some rough instructions on IRC, either the iot channel or in PM, after my meetings tody
14:14:46 <puiterwijk> today*
14:15:28 <pwhalen> puiterwijk, for sure, let me know when you're ready
14:15:38 <pbrobinson> puiterwijk: iot channel would be good
14:15:46 <puiterwijk> Sure
14:15:53 <pwhalen> agreed
14:17:39 <pbrobinson> #topic 3) ==== Fedora 31 ====
14:18:02 <pbrobinson> #info there's a new compose, will be doing another one shortly so please test
14:18:47 <pwhalen> testing the latest compose, looks pretty good. I am seeing some garbled output in uboot on the rpi3, has anyone had a chance to test a recent compose?
14:19:21 <pwhalen> this uboot issue isnt specific to iot of course
14:19:35 <pbrobinson> pwhalen: which release?
14:19:56 <pbrobinson> and garbled on serial console?
14:20:01 <pwhalen> uboot-tools-2020.01-1 I think, didnt verify
14:20:12 <pwhalen> happens on both display and serial
14:20:57 <pwhalen> will verify and file a bz, just wondering if anyone else saw it
14:22:19 <pbrobinson> not that I remember TBH, I'll dig out my rpi3 tomorrow when I'm back near it
14:24:12 <pbrobinson> #topic 4) ==== Open Floor ====
14:24:32 <pbrobinson> does anyone have any general business?
14:25:12 * jsmith has nothing, unfortunately
14:25:58 <puiterwijk> I don't for now
14:26:31 * pbrobinson doesn't either :-/
14:26:37 <pwhalen> ah, did you guys fix the date on the compose?
14:26:47 <pwhalen> noticed it was right
14:27:05 <pbrobinson> no, it seems to be inconsistent
14:27:10 * puiterwijk is going to guess that the compose stuff won the race this time
14:27:13 <pwhalen> yea, ok
14:27:18 <pbrobinson> I was wondering if it was something around cross overs of timezone
14:27:19 <pwhalen> heh
14:27:33 <puiterwijk> pbrobinson: I still think it's the summary file updates
14:27:57 <puiterwijk> i.e. if the summary gets updated between the tree build and the image build pulling down the tree, the dates will be correct. Otherwise the image will have the old tree
14:27:58 <pbrobinson> ah, yea, probably
14:28:44 <pbrobinson> what would update or not update the summary? Does robosingatory do that?
14:28:59 <puiterwijk> Basically, https://kojipkgs.fedoraproject.org/compose/iot/repo/ should not have a summary file
14:29:09 <puiterwijk> robosignatory does not touch the summary file last I touched it
14:29:17 * puiterwijk checks at the latest patches
14:29:47 <puiterwijk> Yeah, it doesn't
14:30:09 <puiterwijk> So that means that something else is triggering the summary file there, which in my opinion is not a file that belongs in the compose repo
14:30:17 <puiterwijk> Because pungi only waits for the actual ref file to be updated, not the summary
14:30:31 <puiterwijk> I have a theory that it's the new-updates-sync script.
14:30:49 <puiterwijk> Or that someone/something else is for some reason building summaries there... I'll go check that out.
14:31:19 <puiterwijk> (Note that the summary file is updated less recent than the refs/ themselves - that's the actual problem)
14:31:44 <pbrobinson> oh, I didn't think we used that for iot yet
14:35:07 <puiterwijk> I might be wrong. Which is why I need to check :)
14:35:16 <puiterwijk> But something is creating the summary file
14:39:17 <puiterwijk> Anyway, I’ll go check that out now/after this meeting
14:39:37 <pbrobinson> thank you all
14:39:37 <pbrobinson> #endmeeting