17:00:00 <mboddu> #startmeeting RELENG (2018-04-05)
17:00:00 <zodbot> Meeting started Thu Apr  5 17:00:00 2018 UTC.  The chair is mboddu. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:00 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:00:00 <zodbot> The meeting name has been set to 'releng_(2018-04-05)'
17:00:00 <mboddu> #meetingname releng
17:00:00 <zodbot> The meeting name has been set to 'releng'
17:00:00 <mboddu> #chair dgilmore nirik tyll sharkcz masta pbrobinson pingou puiterwijk maxamillion mboddu Kellin
17:00:01 <mboddu> #topic init process
17:00:01 <zodbot> Current chairs: Kellin dgilmore masta maxamillion mboddu nirik pbrobinson pingou puiterwijk sharkcz tyll
17:00:43 <mboddu> #topic Roll Call
17:04:07 <mboddu> tik tik tik
17:04:57 <masta> hey there
17:05:40 <Kellin> mboddu: tok tok tok
17:05:48 <mboddu> masta, Kellin : Hello
17:06:56 <mboddu> There are two topics that I want to discuss today which require some historical knowledge, but Patrick is not joining today and Kevin is running late
17:07:18 <mboddu> Kellin and I dont have that historical knowledge :(
17:07:31 <Kellin> four compose, and seven beta ago...
17:07:47 <Kellin> oh wrong kind of 'historical knowledge'
17:07:48 <nirik> morning
17:07:54 <Kellin> allo nirik
17:08:05 <mboddu> nirik: Thanks for joining
17:08:33 <nirik> happy to be here. :)
17:08:38 <mboddu> #topic #7403 Consider no longer producing the Workstation repo tree in composes
17:08:45 <mboddu> #link https://pagure.io/releng/issue/7403
17:09:07 <mboddu> nirik: I got two topics which require your input, and ^ is one of them
17:10:35 <masta> I seem to remember this
17:10:54 <nirik> so, I think we should ask the workstation working group here.
17:11:03 <nirik> releng doesn't decide deliverables or use cases.
17:11:33 <Kellin> I agree
17:12:04 <mboddu> nirik: True, but do we have all the tooling setup to accomplish this?
17:12:15 <nirik> the idea of passing anaconda info for product is interesting... but I would think we would want it implemented before we use it. ;)
17:12:28 <masta> I remember now, it was pxe installs.
17:12:45 <Kellin> mboddu: so when you say tooling setup to acomplish it, are you suggesting we drop the deliverable as suggested in the initial ask?
17:12:48 <nirik> well, if workstation doesn't care about pxe or boot installs that default to workstation we could remove the tree.
17:12:50 * Kellin is still catching up on reading this
17:12:59 <nirik> but thats up to them really.
17:14:04 <nirik> we could file a ticket for them to consider it... if you like I can do so.
17:14:06 <masta> I think there was a ticket for this same ask, in the previous ticket site. It might be informative
17:14:20 <nirik> oh? ok... first I recall it.
17:14:26 <mboddu> nirik: Thanks, that helps
17:14:36 <mboddu> masta: Ohhh
17:15:03 <mboddu> Kellin: Yes, not generating the workstation install tree
17:15:36 <masta> So my knee jerk sentiment is is Everything is what's being used for Workstation, then that could be used for PXE too.
17:15:59 <nirik> it can, but then you have to pick and choose what you want, you don't get it defaulting to what workstation is
17:16:18 <mboddu> masta: Then people has to select what they want
17:16:19 <masta> I'm not sure what's going on with Anaconda
17:16:20 <nirik> if you pxe against the workstation tree, you get exactly what is on the workstation media
17:17:06 <mboddu> masta: If we workstation tree and use Everything for pxe then we should provide a way for users to select whether they want to install workstation or everything
17:17:20 <mboddu> If we remove workstation tree*
17:17:21 <masta> I know it's a smaller use case, but I do mirror fedora archive, and PXE boot a lot around betas
17:17:44 <Kellin> mboddu: it'd also be something places like say, Princeton's Computer Labs might use too
17:17:57 <masta> but I'm not going to be bias, if pxe is not a use case, then so it goes...
17:18:00 <Kellin> we should consult the workstation WG
17:18:12 <nirik> right
17:18:16 <Kellin> it's enough of a use case that we got lots and lots of requests to fix issues with it when I was working on GRUB2
17:18:30 <mboddu> Kellin: Yes, which is what nirik was planning, to file a ticket against Workstation-WG
17:18:44 <Kellin> so is there anything else we can accomplish today?
17:19:48 <mboddu> #info nirik will file a ticket against Workstation-WG and based on the input we get from the WG we will make necessary decisions
17:19:58 <nirik> can do
17:20:23 <mboddu> thanks nirik
17:20:40 <masta> I'd like to track down the alleged previous ticket on this topic.
17:20:49 <masta> so maybe next week
17:21:06 <nirik> masta: let us know if you find it, would be interesting to see the previous discussions
17:22:15 <mboddu> masta: Please let us know if you can find it and I am interested in going through it
17:22:19 <masta> yeah, hopefully I'm not just imagining it.
17:22:39 <mboddu> Okay, moving on
17:22:40 <mboddu> #topic #7398 Make Rawhide repository handling more suitable for automated processing
17:22:46 <mboddu> #link https://pagure.io/releng/issue/7398
17:23:02 <mboddu> masta: haha, how many beers you had today? ;)
17:23:47 <nirik> I think on this one we can close for now and ask him to reopen if the things we have done haven't fixed his use cases enough
17:24:05 <Kellin> nirik: what have we done? (still reading)
17:24:25 <mboddu> nirik: Okay, thanks for all your help on that
17:24:41 * mboddu has to understand mirror-manager more
17:24:52 <nirik> we setup some redirects in mirrormanager
17:24:53 <mboddu> But everyone is saying to stay away from it
17:25:03 <nirik> so it would point f29 (currently) at rawhide
17:25:25 <nirik> it's not so bad...
17:25:28 <Kellin> mboddu: well, we should really understand it, because we have to work with it :)
17:26:11 <mboddu> nirik: I asked Dennis and Patrick few months back and both said to stay away from it :)
17:26:28 <mboddu> s/said/adviced/
17:26:33 <masta> I like the idea, I would hate to thrash any mirrors with a dramatic change in structure.
17:26:38 <mboddu> Kellin: Yeah, we should
17:27:02 * Kellin has to step out to another overlap meeting.  Any big things before I do that?
17:27:17 <nirik> well, some parts of it are tricky for sure, but it's not that complex overall I don't think
17:27:41 <mboddu> Kellin: Nothing important, these two are the one's that I am interested in
17:28:15 <masta> I'd rather hardlink duplicate files in dirs without symbolic links.
17:28:44 <nirik> masta: we can't use symbolic links... we can't be sure all our mirrors support them
17:29:12 <mboddu> masta: For sure I know that all of our mirrors wont support symbolic links
17:30:25 <mboddu> Anyway, closing the ticket for now and ask him to reopen it if he needs anymore help
17:30:32 <nirik> +1
17:30:39 <masta> sure
17:30:56 <mboddu> #info mboddu will close the ticket for now and ask Kamil to reopen the ticket if he needs anymore help
17:32:06 <mboddu> #topic Alternate Architecture Updates
17:32:17 <mboddu> sharkcz: Any updates?
17:34:06 <mboddu> nirik: Is secondary sigul is working again? Sometime yesterday it was having some issue
17:34:35 <nirik> it should be.
17:34:42 <nirik> I restarted it after updates/reboots
17:35:25 <mboddu> nirik: Cool
17:35:41 <mboddu> I guess we dont have anything on alt arches for today
17:35:47 * mboddu hopes everything is fine
17:35:52 <mboddu> #topic Open Floor
17:36:06 <mboddu> nirik: So, how did the update and reboot go?
17:36:16 <nirik> some issues, but there always are. ;)
17:36:21 <mboddu> nirik: I just noticed https://pagure.io/dusty/failed-composes/issue/132#comment-503849
17:36:25 <mboddu> nirik: haha :)
17:36:54 <mboddu> I am guessing its caused by updates/reboots
17:36:56 <nirik> yeah, thats due to those. will fix and close
17:37:05 <mboddu> Cool
17:37:11 <mboddu> nirik++
17:37:47 <mboddu> Anybody has anything else?
17:38:43 <nirik> not off hand
17:38:58 <mboddu> Okay, then I will close the meeting and give 20 min back to you guys
17:39:03 <mboddu> Thank you all for joining
17:39:08 <mboddu> #endmeeting