17:05:23 <mboddu> #startmeeting RELENG (2017-11-02)
17:05:23 <zodbot> Meeting started Thu Nov  2 17:05:23 2017 UTC.  The chair is mboddu. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:05:23 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:05:23 <zodbot> The meeting name has been set to 'releng_(2017-11-02)'
17:05:23 <mboddu> #meetingname releng
17:05:23 <mboddu> #chair dgilmore nirik tyll sharkcz masta pbrobinson pingou puiterwijk maxamillion mboddu Kellin
17:05:23 <mboddu> #topic init process
17:05:23 <zodbot> The meeting name has been set to 'releng'
17:05:23 <zodbot> Current chairs: Kellin dgilmore masta maxamillion mboddu nirik pbrobinson pingou puiterwijk sharkcz tyll
17:05:37 <Kellin> .hello kellin
17:05:37 <nirik> morning
17:05:38 <zodbot> Kellin: kellin 'None' <kellin@retromud.org>
17:05:41 <dustymabe> .hello2
17:05:42 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dustymabe@redhat.com>
17:06:08 * masta waves hello
17:06:27 <maxamillion> .hello2
17:06:28 <zodbot> maxamillion: maxamillion 'Adam Miller' <maxamillion@gmail.com>
17:06:29 <dustymabe> \o
17:06:52 <sharkcz> .hello2
17:06:53 <zodbot> sharkcz: sharkcz 'Dan Horák' <dan@danny.cz>
17:07:51 <mboddu> Okay, lets get started
17:08:12 <mboddu> I want to start with alt arch updates as sharkcz is here and its quick unless he has something :)
17:08:42 <mboddu> #topic Alternate Architectures Updates
17:08:49 <mboddu> sharkcz: any update?
17:09:03 <sharkcz> no change since last week, we are looking good
17:09:28 <mboddu> #info Everything is going good with alternate arches.
17:09:30 <mboddu> Thanks sharkcz
17:09:33 <sharkcz> np
17:12:01 <mboddu> #topic #7113 proposal to run bodhi updates pungi composes with `--no-label`
17:12:14 <mboddu> #link https://pagure.io/releng/issue/7113
17:12:44 <mboddu> So, I wanted to talk to Dennis about it this week, but unfortunately he is not available as he is going through training in Brno
17:12:52 <mboddu> I dont have any update yet
17:13:18 <dustymabe> so threebean did weigh in in the ticket
17:13:19 <mboddu> dustymabe: Do you want to add anything on this topic?
17:13:37 <dustymabe> which was one of the two people we wanted to consult on this topic
17:14:42 <dustymabe> I have sent dennis emails and irc messages trying to get him to review as well
17:14:45 <dustymabe> no success so far
17:15:24 <dustymabe> anybody have any thoughts on that?
17:15:30 <dustymabe> or on the ticket in general?
17:15:43 <mboddu> dustymabe: sure, but Dennis is the one who raised the concerns, we cannot do anything without talking to him
17:15:54 <dustymabe> raised what concerns?
17:16:41 <maxamillion> yeah, I'm still not clear on what the concerns are but I don't want to +1 changing things without understanding the implications
17:17:24 <dustymabe> i'm fine to wait on this one for now, but would like some resolution soon as it affects the versioning of our ostrees that we deliver as updates
17:17:29 * threebean goes to check if we run nightlies with --no-label
17:17:40 <dustymabe> threebean: i'm pretty sure we do
17:18:02 <dustymabe> affects/effects .... blahh!!!
17:18:18 * mboddu also get confused with those two words
17:18:20 <mboddu> :)
17:18:22 <Kellin> can someone summarize what adding the label does/does not do?
17:18:23 <dustymabe> i think i used it write :)
17:18:29 <dustymabe> har har
17:18:49 <Kellin> AI (affects influences), ER (effects are results)
17:18:54 <Kellin> easy way to remember
17:19:04 <dustymabe> Kellin: :)
17:19:23 <maxamillion> oh yeah, we need another definition of what AI and ER mean ... >.>
17:19:28 <threebean> well, I can't find it.  nor can I find an explicit label for the nightlies.
17:19:45 <dustymabe> threebean: according to lsedlar
17:19:52 <dustymabe> from last meeting
17:19:54 <dustymabe> 13:10:05         lsedlar | PDC could import composes without label fine, it already has nighlies without a label
17:20:17 <dustymabe> threebean: could you summarize what adding the label does/does not do?
17:20:25 <dustymabe> mbonnet did that last time for us I believe, he is not here today
17:20:33 <dustymabe> but I can go try to read logs
17:20:37 <mboddu> threebean: well, we dont pass any label(not even --no-label) for nightlies, except saying its a nightly compose by adding --nightly
17:21:24 <threebean> dustymabe: no.  it's not clear to me what the labels are used by.  I think the answer is "nothing".
17:21:37 <threebean> we do use nightlies for the beta and ga composes, run by hand.
17:21:42 <threebean> s/nightlies/labels/
17:21:50 <dustymabe> threebean: yeah
17:21:51 <Kellin> threebean: what do we understand the label to mean when used
17:22:03 <threebean> Kellin: it's not clear to me.
17:22:05 <Kellin> eg: what is an example "label"
17:22:11 <Kellin> when we label, what would a value look like
17:22:16 <threebean> mboddu: ^^
17:22:36 <bowlofeggs> i suggest waiting for dgilmore to weigh in since he had us explicitly make the change to add that flag
17:22:55 <dustymabe> bowlofeggs: yes, as I said earlier it's only been a week for this one, so I'm ok with that
17:23:08 <dustymabe> although he hasn't responded to my requests to weigh in
17:23:09 <mboddu> threebean: we do use labels for beta and final RC composes
17:23:14 <dustymabe> I guess we'll wait for next week
17:23:20 <maxamillion> bowlofeggs: does anyone know why he requested that be added explicitly?
17:24:00 <mboddu> dustymabe: FYI, he is Brno next week as well, but I am planning to talk to him but it depends on his schedule
17:24:07 <mboddu> maxamillion: nope
17:24:08 <bowlofeggs> maxamillion: i believe he told puiterwijk to use it, but i wasn't part of the conversation
17:24:16 <bowlofeggs> maxamillion: so i'm not sure exactly what was said
17:24:17 <maxamillion> alright
17:24:33 <maxamillion> we need to start writing this kind of stuff down somewhere
17:24:38 <mboddu> maxamillion: If anyone knows about it other than Dennis then it would be Patrick
17:24:38 <dustymabe> mboddu: in that case we'll have this same conversation next week
17:24:45 <maxamillion> how long has this been blocked on at this point?
17:24:52 <mboddu> maxamillion: about a week
17:24:56 <dustymabe> maxamillion: since bodhi+pungi became production
17:25:56 <dustymabe> next topic?
17:26:04 <maxamillion> and now we're going to block for another week?
17:26:13 <maxamillion> bleh ... yeah, go ahead to next topic
17:26:30 <dustymabe> maxamillion: unless you know of a way to solve this without dgilmore
17:26:37 <dustymabe> my hands are tied
17:26:58 <maxamillion> dustymabe: I don't, just asking questions to make sure I understand the situation correctly
17:27:05 <dustymabe> yep
17:27:25 <mboddu> #info We are waiting on dgilmore input, so we are going to punt this for next week
17:28:49 <mboddu> Okay next topic, which might be the same output as this one
17:29:05 <mboddu> #topic #7100 new koji tags for atomic host
17:29:13 <mboddu> #link https://pagure.io/releng/issue/7100
17:29:29 * dustymabe waits for people to review
17:31:59 * dustymabe wonders how long to wait
17:33:19 <dustymabe> so this one has been blocked for 2 weeks at this point
17:33:20 <mboddu> dustymabe: I have given my comments on this one
17:33:47 <dustymabe> the question is what are we waiting on dennis for exactly?
17:34:00 * mboddu points to the above topic and Dennis traveling, so cant do much :(
17:34:23 <dustymabe> mboddu: I guess the question is, can any decisions be made without dennis?
17:36:32 <mboddu> dustymabe: Yes. But, he is the only one who knows about releng more than anyone else. So, some topics like this really need his input.
17:36:55 <dustymabe> i would say that mbonnet has pretty good releng knowledge as well
17:37:03 <dustymabe> and he helped me come up with this proposal
17:37:17 <dustymabe> the question is what does this impact if we do implement it?
17:37:25 <dustymabe> atomic host
17:37:33 <dustymabe> anything else?
17:38:21 <maxamillion> mboddu: I think eventually we need to get to a point we're not blocked on him explicitly for the sake of the greater Fedora Community ... there's not a good "vacation factor" here
17:40:04 <dustymabe> yeah. so the reason I want to figure this out is because we need to set up our 'updates' atomic host runs based on this proposal
17:40:29 <dustymabe> which is why I tried to propose it some time before we started making updates runs
17:40:31 <dustymabe> for f27
17:40:48 <dustymabe> if we have to wait another week it may be 'after' f27 is released
17:41:05 <dustymabe> it's whatever, I'm just frustrated
17:41:07 <mboddu> maxamillion, dustymabe : I agree, but we are not there yet, thats all I am saying. We are trying hard to understand all the aspects of releng but it takes time as some of it is mostly historical knowledge on the tool and process
17:41:44 <dustymabe> #info waiting another week for dgilmore's input. the third time is the charm
17:42:08 <maxamillion> fine
17:42:44 <dustymabe> we'll discuss again next week :)
17:43:24 <mboddu> dustymabe: Well, f27 is blocked for another week so we might get it done before f27 release, if it makes any happy :)
17:43:35 <dustymabe> mboddu: i didn'
17:43:42 <dustymabe> didn't realize RC 1.2 didn't pass
17:43:44 <Kellin> dustymabe: to a point, I have an entire epic project right now to map the contents of his brain
17:43:52 <dustymabe> Kellin +1
17:43:54 <Kellin> dustymabe: so we're working on improving the vacation principle
17:44:48 <mboddu> dustymabe: Go/No-Go is happening right now and we have a blocker
17:45:03 <mboddu> thats why I am kinda slow in replying
17:45:51 <maxamillion> oh bummer
17:46:03 <maxamillion> I mean, it seems like it might work out ... but bummer
17:46:29 <dustymabe> yep
17:46:31 <mboddu> maxamillion: yup
17:46:37 <mboddu> So, moving on
17:46:46 <dustymabe> mboddu: yes please. sorry for derailing meeting
17:47:51 <mboddu> dustymabe: no problem, we are sorry as we were able to get you some input on the topic, but we have to work with what we got
17:48:00 <mboddu> #topic Open Floor
17:48:06 <mboddu> Anybody has anything?
17:48:58 <Kellin> just an open request - if you know something about obscure releng scripts please feel free to add the historical info over here: https://taiga.fedorainfracloud.org/project/acarter-fedora-docker-atomic-tooling/us/888
17:49:09 <Kellin> there's a subtask per each script to leave comments on
17:49:47 * Kellin sent this out in a mail to the releng mailing list on Monday
17:50:40 <dustymabe> Kellin: thank you for trying to hunt all this information down
17:51:47 <maxamillion> Kellin: +1
17:51:54 <Kellin> np - also - if you've done compose work, there is a compose SOP update in the queue, please comment on it if you have input: https://pagure.io/releng/pull-request/6733
17:52:08 <mboddu> #info Kellin wants to have input on https://lists.fedoraproject.org/archives/list/rel-eng@lists.fedoraproject.org/thread/5G3T5T5IW3JP3UQ5P5LJSB7X6JO4HLO3/
17:52:15 * Kellin notes all his updates are mostly "if you know something, share so we can all benefit from less single point of failure
17:52:49 <maxamillion> ugh ... waiting on taiga to load ...
17:54:13 <mboddu> Anything else guys?
17:54:53 <Kellin> I'm going to have a process change up on the board today by EOB for composes/syncs, if you have an interest in that, check back on it in the next 12 hours?
17:55:01 <Kellin> mboddu: it's the one we discussed this morning
17:55:41 <maxamillion> Kellin: updated this one, https://taiga.fedorainfracloud.org/project/acarter-fedora-docker-atomic-tooling/task/945 ... is there anything else you were looking for as far as docs for that stuff?
17:55:51 <mboddu> Kellin: hehe, sure
17:56:11 <Kellin> maxamillion: some of them have been "remove this thing", others are used but not in our SOPs
17:56:36 <Kellin> so pretty much if you've gone to do a releng task and found no SOP, tag that script in the docs so I know I need to write one (plus any info you know)
17:57:12 <Kellin> the over-arching epic is top to bottom documenting our whole release process; I have a couple pages of notes on Bodhi from bowlofeggs - and I'm working to do more in between Python3 ports and such for f28
17:57:18 <maxamillion> Kellin: ah ok, I just updated that one linking to the SOP that goes with the script
17:57:28 <maxamillion> anyhoo, I can follow up offline
17:57:38 <Kellin> yeah.  this is basically a full review of all our docs.  I'm also open to ideas on making docs living things so it's less onerous to keep it up to date
17:57:54 * Kellin was thinking like inline docs with sphinx using docstrings or something similar - if you've done that sort of thing, hit me up offline
17:58:05 <Kellin> this is all in nascent stages so input is highly welcome
17:59:19 <maxamillion> yeah, I have
17:59:23 <maxamillion> it's probably a good idea
17:59:48 <maxamillion> we could even setup some stuff that'll extract based on a costom delimiters if we wanted (though, likely not necessary)
18:00:10 <maxamillion> anyhoo ... offline
18:00:28 * Kellin will stop talking now, this is the first time he's had somethign to say and babbled too much :)
18:02:50 <mboddu> #info Kellin will work on inline docs based on docstrings or similar
18:02:57 <mboddu> Okay, is that all?
18:04:15 <mboddu> Okay, everyone, thanks for joining
18:04:20 <mboddu> #endmeeting