15:34:29 #startmeeting RELENG (2015-06-22) 15:34:29 Meeting started Mon Jun 22 15:34:29 2015 UTC. The chair is dgilmore. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:34:29 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:34:44 #meetingname releng 15:34:44 The meeting name has been set to 'releng' 15:34:44 #chair dgilmore nirik tyll sharkcz bochecha masta pbrobinson pingou maxamillion 15:34:44 Current chairs: bochecha dgilmore masta maxamillion nirik pbrobinson pingou sharkcz tyll 15:34:47 #topic init process 15:34:49 who all is here 15:35:08 * sharkcz is here 15:35:20 * maxamillion is here ... sorry for being late 15:36:31 hello 15:36:43 hey there 15:36:54 #topic #6158 Request to discuss Rel-Eng Project Planning Proposal 15:37:01 https://fedorahosted.org/rel-eng/ticket/6158 15:37:13 maxamillion: anything here for us to discuss 15:37:29 or should we remove the meeting keyword until after flock? 15:37:56 dgilmore: yeah, probably after Flock 15:39:21 #topic #6164 bodhi2 status update requested 15:39:28 https://fedorahosted.org/rel-eng/ticket/6164 15:39:38 lmacken: where does bodhi2 stand today? 15:40:24 dgilmore: same spot as it was last week, since we had production explosions last week to deal with 15:40:34 lmacken: okay cheers 15:40:39 Hopefully we can move forward on it this week 15:40:53 that would be good 15:40:55 #topic Secondary Architectures updates 15:40:56 #topic Secondary Architectures update - ppc 15:40:57 yep 15:41:08 pbrobinson: sharkcz: how is ppc this week? 15:41:25 ppc prepping for mass rebuild, I believe we're blocked on a toolchain issue, hopefully fixed soon 15:41:46 other news from this week is EOL of F-20 which will also be EOL of 32bit ppc 15:41:58 that's about it for power 15:42:32 ack, I see f23 builds running, so we are getting closer 15:43:56 #topic Secondary Architectures update - s390 15:44:02 sharkcz: how is s390? 15:45:04 looks good, we were doing builds from Friday on last Friday (= almost on par with primary), but processing the mass rebuild now 15:45:15 okay cool 15:45:34 #topic Secondary Architectures update - arm 15:45:44 so we're processing the mass rebuild now too 15:45:56 cool. does it seem to be going well? 15:46:03 we're about 9K builds behind but most of them are blocked due to a toolchain issue 15:46:16 most of what was done to date was noarch imports 15:46:19 hrrm okay 15:46:45 the toolchain team had managed to reproduce on Friday so I'm hoping for a fix soon 15:47:20 that would be awesome 15:47:21 in other news it looks like the ghc issue is fixed and due to the mass rebuild we'll come out the other side with all of that in a much better state than it has been 15:47:39 does not seem great we have tool chain issues on two different arches 15:48:27 dgilmore: there seems to be regular toolchain issues across all arches, there will no doubt be a few more come through given it's the first mass rebuild with 5.x 15:49:28 anyway 15:49:38 #topic Open Floor 15:50:09 I will be sparodically working this week as I go to FUDCon APAC 15:50:27 I'll be at Summit starting tomorrow morning and will mostly be AFK through out the week 15:50:43 I did want to bring this up in Open Floor just as something to be on people's radar --> https://fedoraproject.org/wiki/Changes/Layered_Docker_Image_Build_Service 15:50:47 maxamillion: by AFK you mean drunk right? 15:50:53 maxamillion: :) 15:51:02 I had a few items to note from this weekend... 15:51:02 pbrobinson: negative 15:51:08 I have one - someone please untag builds from f22-updates-testing that went into f22 GA, there are scripts in the releng repo to find the NVRs 15:51:12 we also need to get stuck into pungi work 15:51:32 since we are about a month from branching for ALpha 15:51:33 unless already done, haven't checked now 15:51:34 sharkcz: yes, that's on my todo list 15:51:41 pbrobinson: thx 15:51:48 rawhide compose broke due to a createrepo bug. I untagged the new createrepo and it finished. 15:51:57 dgilmore: we need to align with RCM and share that load, I think we're spinning wheels in mud on things that I imagine dmach would know how to fix because he wrote it 15:52:05 nirik: is that createrepo_c or just creatrepo? 15:52:26 I have been signing rawhide after the mass rebuild. I've done about 40k of the 100k so far. ;) 15:52:33 pbrobinson: createrepo. 15:52:43 as far as I know we haven't switched to createrepo_c in mash yet 15:53:17 maxamillion: we do need to work with them 15:53:24 dgilmore: +1 15:53:27 maxamillion: but we have a hard deadline to get things working 15:53:55 nirik: from memory and looking at the history its been broken for awhile 15:54:03 dgilmore: rgr 15:54:17 and we have untagged newer builds for breaking things 15:54:21 dgilmore: it was broken before, but that build never worked/pushed out... so the mass rebuild picked it up when it rebult 15:54:26 yeah 15:54:41 anyhow, I don't care if we switch to createrepo_c or fix this or whatever, just letting people kow 15:54:44 know even 15:55:03 nirik: I do want to get switched to createrepo_c 15:55:11 maybe this is the omph to do so 15:55:44 oh, and FYI, I will be gone all next week on PTO in the mountains. 15:55:56 nirik: :) nice 15:56:06 seems many people will be AFK then 15:56:08 right, so next week looks like we'll be a bit thin on the ground 15:56:10 nirik: sounds great, have fun and safe travels! 15:56:14 I will be mostly AFK next week also 15:56:18 will have some net, but will try not to use it. ;) 15:56:29 nirik: yeah, don't .... like not even a little 15:56:31 nirik: good! 15:56:38 nirik: UNPLUG MAN 15:56:50 do we need to reorg the push duty? I can't recall who is on... 15:56:56 .pushduty 15:57:10 I'll be about, unlike others I don't have an interesting travel plan next week ;-) 15:57:29 nirik: The following people are on push duty: Fedora Release Engineering, Jon 15:57:32 nirik: - https://apps.fedoraproject.org/calendar/release-engineering/ 15:58:04 I'm next week. 15:58:12 masta: push as late as possible tonight 15:58:13 masta: you want to trade? I take this week, you take next? 15:58:20 masta: since it is the last f20 push 15:58:26 or that 15:58:32 nirik: ^^^^^^^^^^^^^^^^^ 15:58:39 sure, we can sort it out. 15:58:44 last f20 push is this week isn't it? 15:58:51 yep. today even. ;) 15:59:07 nirik: I could probably do it if someone can help me get up to speed this week? 15:59:09 f20 EOL notice goes out tomorrow 15:59:11 dgilmore: hey 15:59:21 pbrobinson: I can show you on Wednesday 15:59:38 nirik: and then sched me into the roster 15:59:43 so I have a push going already 15:59:50 is that going to be a problem? 15:59:55 yes 15:59:58 well kinda 16:00:06 ok 16:00:19 I guess we can try to stop the push that is processing. 16:00:29 masta: ehh it is done 16:00:30 sry about that, did that first thing this morning 16:00:42 you know, try to knock those out early 16:00:50 my bad =( 16:00:53 maybe it will finish quickly and we can do another to get the last minute straglers 16:01:06 masta: EOL day is a special one 16:01:30 dgilmore: yeah, i think it will be a quick one 16:01:37 not too many thing in the push 16:01:43 * nirik did pushes saturday and sunday. 16:01:54 had to do one saturday to try and get mm back happy 16:02:07 because we had all kinds of doom thursday->saturday morning. :( 16:02:30 nirik: were you going to give a quick overview of what happened there? 16:02:48 dgilmore: what should I do? 16:03:47 well, we don't know the cause really, but short story: large network burst caused virthost02 to act all wacky/slow. That cause our main db server and vpn hub to be slow and caused all kinds of problems. Rebooting the vm's didn't help, but rebooting the entire virthost got it back to normal. 16:04:15 and there may be a mirrormanager bug around alternative repomd's 16:04:19 * nirik should file that. 16:04:37 masta: let it run 16:07:25 does anyone have anything else? 16:08:02 * dgilmore will take that as a no 16:08:06 #endmeeting