From Fedora Project Wiki

Fedora Release Engineering Meeting :: Monday 30-APR-07

Reports on F7-Test 4

Merge

  • Decisions
  • Buildsys freeze on 01-MAY-07
  • Start merge on 02-MAY-07 at 12:00 EDT until completed
  • move final devel freeze out one week
  • hold current GA date of 24-MAY-07
  • http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge
  • http://fedoraproject.org/wiki/Infrastructure/CoreExtrasCVSMerge should be pretty detailed for the cvs side of the merge
  • f13
  • I'm continuing to keep koji updated with the Core builds
  • dgilmore had something cooking to import Extras builds, we should probably start doing that regularly ASAP
  • the question marks are, a owners.list <-> koji sync tool
  • need the creation of a needsign type que for up to the hour build access and a way to do package signing.
  • how did we wind up with a 1~ week window between Test4 release and Final freeze?
  • moving ahead with hardware we have; new hardware is on the way
  • it could take a while to import the Extras stuff becasue I'll have to write something up to do the imports since Extras isn't in a koji like builddb right now
  • we will let people do post-merge rebuilds to pick up new builddeps
  • there is also a ton of dist-fc7 builds that haven't been requested for f7-final tagging.
  • we will disable the build system by removing the dist-fc7 build target from brew.
  • koji will first be used for f7 only and after the dust settles we'll bring over fc6-updates
  • will document the freeze process
  • will send mail on freeze
  • jeremy
  • people will still have to request their post-merge rebuilds be tagged for final
  • mmcgrath
  • will try to clean up the filesystem on cvs too before next wed. It's kind of a mess on that box right now.
  • new PPC arriving late this week or beginning of next
  • notting
  • when can we reasonably have a deadline?
  • will create the hook for the owners.list <-> koji sync
  • poelcat
  • suggested detailed schedule with completion dates for merge
  • others thought a list of tasks is fine because everything has to be done on Wednesday

IRC Transcript

* notting looks around<a href="#t13:01" class="time">13:01</a>
vigneshnman64, Hey<a href="#t13:02" class="time">13:02</a>
* jeremy is here<a href="#t13:02" class="time">13:02</a>
jeremyf13 just sent me a message saying he'd be a little late.  but also that he's not sure there's anything in particular that needs discussing<a href="#t13:02" class="time">13:02</a>
vigneshwhen does nman64 come around here ?<a href="#t13:02" class="time">13:02</a>
nottingum, *merge*?<a href="#t13:02" class="time">13:02</a>
jeremynotting: is that release team, though?   (granted, lots of overlap)<a href="#t13:03" class="time">13:03</a>
rdieterhere<a href="#t13:03" class="time">13:03</a>
nottingjeremy: sort of<a href="#t13:04" class="time">13:04</a>
rdieterkde -> Extras?  :)<a href="#t13:04" class="time">13:04</a>
wwoodsthe release is kind of gated by the merge<a href="#t13:04" class="time">13:04</a>
wwoodsalso it looms overhead and gives me nightmares<a href="#t13:05" class="time">13:05</a>
vigneshwill the new udev be in Fedora 7 ?<a href="#t13:05" class="time">13:05</a>
jeremyrdieter: I think we're looking at more generally doing everything instead of just doing kde...<a href="#t13:05" class="time">13:05</a>
vigneshbecause the 2.6.20 kernel ata as sata too..  and doesn't boot<a href="#t13:05" class="time">13:05</a>
rdieteryou guys keep saying that, and that's fine, as long as it happens *real fast*.  :)<a href="#t13:06" class="time">13:06</a>
jeremynotting: so, we need to a) schedule the downtime b) do the bits to make it happen c) wait and suffer as builds take a lot longer ;-)<a href="#t13:07" class="time">13:07</a>
nottingjeremy: who has the ability to do all these things?<a href="#t13:07" class="time">13:07</a>
nottingdamn, 532 messages in fedora-*<a href="#t13:08" class="time">13:08</a>
jeremynotting: the first needs coordination with infrastructure.  the second requires f13 for koji probably and somebody to do the cvs side.  and the third is everyone! :)<a href="#t13:08" class="time">13:08</a>
jeremyand yeah, way too much mail this weekend<a href="#t13:09" class="time">13:09</a>
nottingjeremy: so do we have enough people here to make a plan?<a href="#t13:11" class="time">13:11</a>
jeremynotting: would help if we had f13 and mmcgrath.  f13 says he'll be here in a few<a href="#t13:12" class="time">13:12</a>
nottingfailing that, any current release concerns?<a href="#t13:12" class="time">13:12</a>
nottinghas someone swept the queue of things built internally but not requested for f7-final?<a href="#t13:12" class="time">13:12</a>
warrenIs qemu-kvm working for anybody with the latest rawhide?<a href="#t13:13" class="time">13:13</a>
jeremyno, was going to try to whip up a script to run daily and send out a list<a href="#t13:13" class="time">13:13</a>
nottingwarren: -ENOHW<a href="#t13:14" class="time">13:14</a>
jeremywarren: haven't rebooted yet today, but working fine and I don't think I've seen anything relevant go in anyway<a href="#t13:14" class="time">13:14</a>
warrenIt was working up until the last two kernels for me.<a href="#t13:14" class="time">13:14</a>
warrenSorry, I don't mean this to be off-topic, notting asked about release concerns.<a href="#t13:15" class="time">13:15</a>
notting*shrug* broken features are release concerns<a href="#t13:16" class="time">13:16</a>
wwoodsresume regressed on my thinkpad :/ but that's something I'll track down and report in a bit<a href="#t13:16" class="time">13:16</a>
nottingwwoods: automated on rawhide still happy?<a href="#t13:16" class="time">13:16</a>
jeremywarren: I'll reboot in a bit and make sure<a href="#t13:16" class="time">13:16</a>
wwoodsactually no<a href="#t13:17" class="time">13:17</a>
wwoodstoday's doom-o-meter: 100%<a href="#t13:17" class="time">13:17</a>
warrenjeremy, using x86_64 kernel?<a href="#t13:17" class="time">13:17</a>
jeremywarren: my workstation is<a href="#t13:17" class="time">13:17</a>
wwoodsit hit 100% after Friday's changes<a href="#t13:17" class="time">13:17</a>
wwoodsso rawhide is currently uninstallable<a href="#t13:17" class="time">13:17</a>
nottingwwoods, ok, that sounds bad<a href="#t13:18" class="time">13:18</a>
wwoodstraceback in yum<a href="#t13:18" class="time">13:18</a>
jeremywwoods: I think I know what's causing it and have already fixed it.  I can get you an updates.i.mg to try<a href="#t13:18" class="time">13:18</a>
wwoodsjeremy: would appreciate that<a href="#t13:18" class="time">13:18</a>
jeremy(basically, we reverted an API change in yum so that yum 3.2 could be compatible with yum 3.0.x...  but forgot to revert the anaconda part)<a href="#t13:18" class="time">13:18</a>
wwoodsI'm really concerned about mkinitrd issues - there's a lot of "can't find /dev/root" reports out there<a href="#t13:19" class="time">13:19</a>
nottingwwoods, example?<a href="#t13:19" class="time">13:19</a>
wwoodsfurthermore we're gonna need some docs on device labeling, since that's a very common install failure<a href="#t13:19" class="time">13:19</a>
warrenwwoods, any commonality in the reporters?<a href="#t13:19" class="time">13:19</a>
jeremywwoods: upgrade you mean?  (for device labeling; for installs, it should Just Happen (tm))<a href="#t13:20" class="time">13:20</a>
wwoodsI might just be confusing some livecd/SCSI issues as well<a href="#t13:21" class="time">13:21</a>
wwoodsjeremy: err, yeah, upgrade<a href="#t13:21" class="time">13:21</a>
jeremythe idea was going to be that some docs would be written... probably hasn't actually happened yet, though (for labels + upgrades)<a href="#t13:22" class="time">13:22</a>
jeremyas for /dev/root stuff...  wwoods: if you find bugs, you can add me to the cc and I can take a look at them<a href="#t13:22" class="time">13:22</a>
jeremythat way we can at least try to have some sort of triage/common-thread-finding on them<a href="#t13:22" class="time">13:22</a>
warrenjeremy, <a href="http://people.redhat.com/wtogami/temp/qemu-kvm-failure.txt">http://people.redhat.com/wtogami/temp/qemu-kvm-failure.txt</a>  is this proper looking syntax for qemu-kvm?<a href="#t13:22" class="time">13:22</a>
wwoods<a href="https://www.redhat.com/archives/fedora-test-list/2007-April/msg00659.html">https://www.redhat.com/archives/fedora-test-list/2007-April/msg00659.html</a> is a good place to start<a href="#t13:23" class="time">13:23</a>
jeremywarren: yes.  I have a hunch as to what that might be even.  stay tuned :)<a href="#t13:23" class="time">13:23</a>
jeremywwoods: yeah, might even be worth asking the guy if he's willing to write up a first draft and it can at least be in the web release notes<a href="#t13:24" class="time">13:24</a>
wwoods<a href="https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=234938">https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=234938</a> is a failed upgrade<a href="#t13:24" class="time">13:24</a>
wwoods<a href="https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=236743">https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=236743</a><a href="#t13:24" class="time">13:24</a>
wwoodsbut those are anaconda, not mkinitrd<a href="#t13:24" class="time">13:24</a>
jeremywwoods: either setting something in their whiteboard or listing them on a wiki so we can easily get back to them is going to be the best thing to do.  just listing them here isn't going to go anywhere fast :-/    (well, we _COULD_ sit and make this take a few hours while I try to get to looking at them, but... :-)<a href="#t13:26" class="time">13:26</a>
wwoodsI was thinking of doing a mkinitrd tracker bug, but QA Whiteboard is probably better<a href="#t13:26" class="time">13:26</a>
wwoodsI'll do another one for the anaconda partitioning stuff<a href="#t13:27" class="time">13:27</a>
wwoodsprobably I'll start adding those anaconda failures to the blocker list<a href="#t13:27" class="time">13:27</a>
wwoodslet me know if I get overzealous with adding blockers and such<a href="#t13:27" class="time">13:27</a>
jeremyI'd rather have them where we can find them easily and then throw them off than lose them in the massive noise that is everything filed<a href="#t13:28" class="time">13:28</a>
jeremyso as long as every bug that gets filed isn't ending up there, we're probably okay :-)<a href="#t13:28" class="time">13:28</a>
f13Ok, I'm here.<a href="#t13:31" class="time">13:31</a>
f13jeremy: notting: well, this is the weekly release-team meeting too, it just is convienent to have the Release Meeting at hte same time when there are releases coming up<a href="#t13:32" class="time">13:32</a>
jeremyso, merge<a href="#t13:34" class="time">13:34</a>
f13merge.<a href="#t13:35" class="time">13:35</a>
f13I'm continuing to keep koji updated with the Core builds<a href="#t13:35" class="time">13:35</a>
f13dgilmore had something cooking to import Extras builds, we should probably start doing that regularly ASAP<a href="#t13:35" class="time">13:35</a>
jeremy*nod*<a href="#t13:37" class="time">13:37</a>
f13I think the question marks are, a owners.list <-> koji sync tool<a href="#t13:37" class="time">13:37</a>
nottingwhen can we reasonably have a deadline?<a href="#t13:37" class="time">13:37</a>
f13the creation of a needsign<a href="#t13:37" class="time">13:37</a>
f13type que for up to the hour build access<a href="#t13:38" class="time">13:38</a>
wwoodswho owns the owners.list <-> koji sync bit?<a href="#t13:38" class="time">13:38</a>
f13and a way to do package signing.<a href="#t13:38" class="time">13:38</a>
f13wwoods: nobody yet, it's listed as ? in <a href="http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge">http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge</a><a href="#t13:38" class="time">13:38</a>
jeremyadded fun: after tomorrow, my availability may be sporadic for the rest of the week and then I'm in san diego next week.<a href="#t13:38" class="time">13:38</a>
wwoodsI'm unfamiliar with both halves but if it helps, I'll take it<a href="#t13:38" class="time">13:38</a>
nottingjeremy, so, this week or bust?<a href="#t13:38" class="time">13:38</a>
jeremynotting: I think we pretty much ahve to do this week, yeah<a href="#t13:39" class="time">13:39</a>
f13yeah<a href="#t13:39" class="time">13:39</a>
f13how hte heck did we wind up with a 1~ week window between Test4 release and Final freeze?<a href="#t13:39" class="time">13:39</a>
f13doing the merge and having final devel freeze of... Thursday, as in 3 days from now... yikes.<a href="#t13:40" class="time">13:40</a>
wwoodsI think we assumed the merge would be done by now.<a href="#t13:40" class="time">13:40</a>
jeremywwoods: that's one part, yes<a href="#t13:41" class="time">13:41</a>
f13nod.<a href="#t13:41" class="time">13:41</a>
jeremyf13: part of it was "cut down on spurious changes between test4 and final"<a href="#t13:41" class="time">13:41</a>
wwoodsare there any plans for doing the comps-diff dance to figure out what extra software should be installed during upgrade?<a href="#t13:41" class="time">13:41</a>
f13and having the summit doesn't help.<a href="#t13:41" class="time">13:41</a>
poelcatI think having a more detailed schedule could help<a href="#t13:41" class="time">13:41</a>
wwoodsbecause I'm manually installing gnome-power-manager and some other stuff and wishing it just happened automatically. again.<a href="#t13:42" class="time">13:42</a>
poelcatat least we would all be more aware of what some of the detailed steps are and what we have to work around<a href="#t13:42" class="time">13:42</a>
wwoodsactually. probably this is an F8 anaconda feature. so nevermind.<a href="#t13:42" class="time">13:42</a>
f13wwoods: have we ever done "extra" software installs during upgrades?  I thought upgrades were purely that, just upgrades, no extra stuff thrown in by surprise.<a href="#t13:43" class="time">13:43</a>
wwoodsyeah. if you ask me, that's the wrong behavior.<a href="#t13:43" class="time">13:43</a>
jeremywwoods: and if I ask 3 other people, I get three different answers.  but discussion for a different time and place<a href="#t13:43" class="time">13:43</a>
wwoodsagreed<a href="#t13:43" class="time">13:43</a>
f13poelcat: suggestions?  <a href="http://fedoraproject.org/wiki/Releases/7/Schedule">http://fedoraproject.org/wiki/Releases/7/Schedule</a> and <a href="http://fedoraproject.org/wiki/ReleaseEngineering/Overview">http://fedoraproject.org/wiki/ReleaseEngineering/Overview</a><a href="#t13:43" class="time">13:43</a>
poelcatf13: add the summit to the schedule; add interim milestones; if we don't hit those, then we'll know earlier we won't hit the end one<a href="#t13:45" class="time">13:45</a>
f13so here's the thing.  we need a bit of an outage to complete the merge.  Is it prudent to do such an outage at this time, or should we kick out the final devel freeze a week?<a href="#t13:45" class="time">13:45</a>
f13poelcat: good suggestions.  can you remember then when we draft up the schedule for 8?  (:<a href="#t13:45" class="time">13:45</a>
poelcatf13: right now all we have is "do the merge"<a href="#t13:45" class="time">13:45</a>
poelcatbut no timeline or what steps that requires<a href="#t13:46" class="time">13:46</a>
nottingf13, if the outage is going to encompass the entire time between now and the freeze, that might not be good<a href="#t13:46" class="time">13:46</a>
poelcatso how do we know how far we are in "doing the merge?"<a href="#t13:46" class="time">13:46</a>
nottingf13, url for merge planning page?<a href="#t13:46" class="time">13:46</a>
nottingf13, nm<a href="#t13:46" class="time">13:46</a>
f13<a href="http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge">http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge</a><a href="#t13:47" class="time">13:47</a>
f13poelcat: see above.  I tried to update it with all i know.<a href="#t13:47" class="time">13:47</a>
warrenDoes this mean we have sufficient build hardware now?  Or we're just going ahead anyway?<a href="#t13:48" class="time">13:48</a>
jeremypoelcat: in some ways, F7 is a special case _because_ of the merge... it's not really a release task, but it has a large impact on the release<a href="#t13:48" class="time">13:48</a>
poelcatf13: i'll look... too bad I didn't think of talking about this stuff when we were at the NWLinux conf :(<a href="#t13:48" class="time">13:48</a>
poelcatjeremy: every release is "special" :)<a href="#t13:48" class="time">13:48</a>
nottingf13: do we have a packagedb <-> koji sync?<a href="#t13:48" class="time">13:48</a>
jeremynotting: packagedb isn't production yet afaik...<a href="#t13:48" class="time">13:48</a>
nottingjeremy, might as well break everything!<a href="#t13:49" class="time">13:49</a>
f13poelcat: yeah, things were really busy then though.<a href="#t13:50" class="time">13:50</a>
poelcattrue<a href="#t13:50" class="time">13:50</a>
f13wwoods: we're going ahead, hardware is on order on the way<a href="#t13:50" class="time">13:50</a>
nottingf13, more seriously, from what i understand of koji/brew, a dumb owner sync tool would be pretty trivial<a href="#t13:51" class="time">13:51</a>
f13poelcat: it was less crazy this year since it was 2 days, but still a lot going on<a href="#t13:51" class="time">13:51</a>
f13notting: it should be, somebody just has to write it.<a href="#t13:51" class="time">13:51</a>
nottingf13, so, '<a href="#t13:51" class="time">13:51</a>
nottingf13, so, 'friday morning didn't happen'<a href="#t13:52" class="time">13:52</a>
nottingwhen do we want to schedule?<a href="#t13:52" class="time">13:52</a>
f13Tonight is bad, I'm on a plane.  Tomorrow night might be bad, I'll likely be a zombie from doing a red-eye and then a work day.  Wed is the day before the freeze....<a href="#t13:52" class="time">13:52</a>
nottinghow long does it take to import?<a href="#t13:53" class="time">13:53</a>
f13notting: it could take a while to import the Extras stuff<a href="#t13:53" class="time">13:53</a>
nottingcan we start now, then?<a href="#t13:54" class="time">13:54</a>
f13dgilmore: ping?<a href="#t13:54" class="time">13:54</a>
f13If he doesn't have anything, I'll have to write something up to do the imports.<a href="#t13:54" class="time">13:54</a>
jeremydgilmore might be in .au right now, not 100% sure<a href="#t13:55" class="time">13:55</a>
f13since Extras isn't in a koji like builddb right now, it'll be a bit difficult to manage, but I'll figure something out.<a href="#t13:55" class="time">13:55</a>
f13nod<a href="#t13:55" class="time">13:55</a>
nottingwell, ithink we need to a) merge this week b) move final freeze<a href="#t13:55" class="time">13:55</a>
nottingand we should start importing builds now, and set a buildsys freeze date today and send mail<a href="#t13:56" class="time">13:56</a>
wwoodsagreed, but does pusing the freeze push the release?<a href="#t13:56" class="time">13:56</a>
nottingwhether that date is wednesday or not, i dunno<a href="#t13:56" class="time">13:56</a>
rdieternotting: +1  (selfishly speaking, I'll need more time for kde work).<a href="#t13:56" class="time">13:56</a>
warrenrdieter, you need to do it in plague?<a href="#t13:56" class="time">13:56</a>
jeremywwoods: potentially<a href="#t13:56" class="time">13:56</a>
nottingrdieter, what needs done besides rebuild with --with-frobozz?<a href="#t13:57" class="time">13:57</a>
f13the merge will destabalize things which sucks, as the tree is in pretty good shape now. So it could push out the final release date a bit, but since we missed the Summit, tha't snot hte end of the world.<a href="#t13:57" class="time">13:57</a>
rdieternotting: that's most of it, + a few package refactoring, splits.<a href="#t13:57" class="time">13:57</a>
wwoodsare we still letting people do post-merge rebuilds to pick up new builddeps?<a href="#t13:58" class="time">13:58</a>
wwoods(i.e. OH NOES! further destabilization!)<a href="#t13:58" class="time">13:58</a>
f13wwoods: that was the plan :/<a href="#t13:58" class="time">13:58</a>
f13there is also a ton of dist-fc7 builds that haven't been requested for f7-final tagging.<a href="#t13:59" class="time">13:59</a>
wwoodspeople will still have to request their post-merge rebuilds be tagged for final, though, right?<a href="#t13:59" class="time">13:59</a>
jeremywwoods: yes<a href="#t13:59" class="time">13:59</a>
nottingf13, wednesday should give enough time to script up a owners -> koji, and some cvs-fu<a href="#t14:00" class="time">14:00</a>
nottingf13, what's involved in disabling the build system?<a href="#t14:00" class="time">14:00</a>
f13So, start merger on Wed, push final freeze to Next Thursday?  (well, that's the summit, so Tuesday after that?)<a href="#t14:01" class="time">14:01</a>
nottingis *everyone* at the summit?<a href="#t14:01" class="time">14:01</a>
f13notting: removing the dist-fc7 build target from brew.<a href="#t14:01" class="time">14:01</a>
nottingf13, and for extras?<a href="#t14:01" class="time">14:01</a>
f13Not sure how to disable the build target there, but I assume it is somewhat the same.<a href="#t14:01" class="time">14:01</a>
f13CVS could be locked out too for those branches right?  Prevent tagging from happening?<a href="#t14:01" class="time">14:01</a>
nottingok, so need to track down dgilmore for that<a href="#t14:02" class="time">14:02</a>
mmcgrathf13: cvs could be locked out pretty easily.<a href="#t14:02" class="time">14:02</a>
jeremynotting: you're not!<a href="#t14:02" class="time">14:02</a>
jeremynotting: everyone else, ... :)<a href="#t14:02" class="time">14:02</a>
nottingjeremy, rdieter? dgilmore? jwb?<a href="#t14:02" class="time">14:02</a>
nottingf13, what releases are we using koji for? f7 only? epel5? ??<a href="#t14:02" class="time">14:02</a>
jeremyyeah, cvs lock down is easy<a href="#t14:02" class="time">14:02</a>
rdieternotting: summit?  yes.<a href="#t14:03" class="time">14:03</a>
jeremynotting: jwb isn't.  don't think dgilmore is either (since I think he's in australia still :)<a href="#t14:03" class="time">14:03</a>
warrennotting, I wont be at the summit.<a href="#t14:03" class="time">14:03</a>
f13notting: First would be f7 only.<a href="#t14:03" class="time">14:03</a>
jwbhuh?<a href="#t14:03" class="time">14:03</a>
f13notting: after the dust settles we'll bring over fc6-updates<a href="#t14:03" class="time">14:03</a>
nottingf13, so, if you document the freeze process, we can still freeze next thursday<a href="#t14:04" class="time">14:04</a>
f13koji is synced up to core as of 5 minutes ago.<a href="#t14:04" class="time">14:04</a>
* jwb is in Real Work Mode<a href="#t14:04" class="time">14:04</a>
warrenf13, synced to f7-final or dist-fc7?<a href="#t14:04" class="time">14:04</a>
jeremyjwb: just saying you're not going to be at the summit<a href="#t14:04" class="time">14:04</a>
f13notting: theoretically I could do it myself too, there will be wireless I assume.<a href="#t14:04" class="time">14:04</a>
f13actually<a href="#t14:04" class="time">14:04</a>
f13the final freeze doesn't need any action<a href="#t14:04" class="time">14:04</a>
nottinghow much cvs do we want to merge?<a href="#t14:04" class="time">14:04</a>
nottingdevel? fc6/fc5?<a href="#t14:04" class="time">14:04</a>
f13we're still using f7-final to do composes from, we just stop processing tag requests, or get a whole lot more strict on what we'll allow to be tagged.<a href="#t14:05" class="time">14:05</a>
nottingf13, yeah, but a quicko 'these are the settings i use for pungi' doc<a href="#t14:05" class="time">14:05</a>
f13ah right<a href="#t14:05" class="time">14:05</a>
nottingand we can fire off some test trees<a href="#t14:05" class="time">14:05</a>
jeremynotting: the plan a little while ago was devel (f7) only at first<a href="#t14:05" class="time">14:05</a>
f13I should have all that ready to go before I leave for the summit<a href="#t14:05" class="time">14:05</a>
f13/should/<a href="#t14:05" class="time">14:05</a>
jwbjeremy, unfortunately, you're correct<a href="#t14:06" class="time">14:06</a>
f13We'll want the rpm that is in updates-testing to go to updates to fix some build problems, or we'll just use the -testing rpm<a href="#t14:06" class="time">14:06</a>
nottingjeremy, ok, easy enough<a href="#t14:06" class="time">14:06</a>
jeremynotting: <a href="http://fedoraproject.org/wiki/Infrastructure/CoreExtrasCVSMerge">http://fedoraproject.org/wiki/Infrastructure/CoreExtrasCVSMerge</a> should be pretty detailed for the cvs side of the merge<a href="#t14:07" class="time">14:07</a>
jeremynotting: couple of open questions, but it was pretty close to ready to go when we bailed last time<a href="#t14:07" class="time">14:07</a>
warrenjeremy, mind if I add dead.package detection?<a href="#t14:07" class="time">14:07</a>
nottingjeremy, why do for /cvs/dist/rpms, rather than just generate the list from rawhide?<a href="#t14:08" class="time">14:08</a>
jeremywarren: go for it.  it's open to help<a href="#t14:08" class="time">14:08</a>
f13mbonnet: other than rpm, can you think of any blockers to starting the merger this Wed?<a href="#t14:08" class="time">14:08</a>
jeremynotting: easier scripting.  either way works<a href="#t14:08" class="time">14:08</a>
mbonnetf13: we happy with the state of the buildsys hardware?<a href="#t14:09" class="time">14:09</a>
f13mbonnet: with two more machines ordered and potentially on the way, yes.<a href="#t14:09" class="time">14:09</a>
f13mmcgrath: heard anything about the machines on order yet?<a href="#t14:09" class="time">14:09</a>
mmcgrathYeah, I got an email this morning, the wording was odd.  Let me dig it up.<a href="#t14:09" class="time">14:09</a>
mbonnetf13: other than that, no, I think we're in good shape<a href="#t14:10" class="time">14:10</a>
mmcgrathLooks like late this week or early next week for the P505 (PPC)<a href="#t14:10" class="time">14:10</a>
f13goot.<a href="#t14:11" class="time">14:11</a>
nottingso, pick a time?<a href="#t14:15" class="time">14:15</a>
nottingwednesday at noon?<a href="#t14:15" class="time">14:15</a>
nottingto be unfrozen 'when it's done'?<a href="#t14:15" class="time">14:15</a>
f13sure!<a href="#t14:15" class="time">14:15</a>
nottingi am ASSuming that our koji install has been tested to do something sane when asked to build? :)<a href="#t14:16" class="time">14:16</a>
rdietershowdown at the OK Fedora coral, Wed, high noon.<a href="#t14:16" class="time">14:16</a>
mmcgrathis the koji portion blocking on dgilmore?<a href="#t14:16" class="time">14:16</a>
* mmcgrath doesn't think it is but is making sure.<a href="#t14:16" class="time">14:16</a>
nottingf13, can you send mail? :)<a href="#t14:17" class="time">14:17</a>
notting(i am somewhat mail-less ATM)<a href="#t14:17" class="time">14:17</a>
f13mmcgrath: not really.  I'd like to have what he was planning for Extras import, but I can cook something up myself.<a href="#t14:18" class="time">14:18</a>
* mmcgrath notes ab is here if we want to talk about pkgdb.<a href="#t14:18" class="time">14:18</a>
mmcgratherr abadger1999<a href="#t14:18" class="time">14:18</a>
f13notting: We've been testing koji out a lot, doing lots of build tests and such.<a href="#t14:18" class="time">14:18</a>
f13notting: yeah, I'll send the mail-o-doom.<a href="#t14:19" class="time">14:19</a>
wwoodsprepare kittens for blending!<a href="#t14:19" class="time">14:19</a>
* abadger1999 waves<a href="#t14:19" class="time">14:19</a>
* notting watches his laptop crash *AGAIN*<a href="#t14:20" class="time">14:20</a>
f13oh fun<a href="#t14:20" class="time">14:20</a>
f13Ok, anything else on the merger?  Anybody want to be on the hook for the owners.list <-> koji sync?<a href="#t14:21" class="time">14:21</a>
nottingf13, sure, i'll do it<a href="#t14:23" class="time">14:23</a>
nottingf13, it's going to be a dumb-and-slow-but-correct script<a href="#t14:23" class="time">14:23</a>
mmcgrathI'll try to clean up the filesystem on cvs too before next wed.  It's kind of a mess on that box right now.<a href="#t14:24" class="time">14:24</a>
f13notting: thanks, please mark your space on the wiki page.<a href="#t14:25" class="time">14:25</a>
nottingdone<a href="#t14:28" class="time">14:28</a>
f13Anything else or can we adjurne, to meet again Wed before noon?<a href="#t14:28" class="time">14:28</a>
wwoodswhen on wed.? "before noon" seems a bit broad.<a href="#t14:29" class="time">14:29</a>
warren11AM EDT?<a href="#t14:30" class="time">14:30</a>
f13Sure, why not.<a href="#t14:31" class="time">14:31</a>
poelcatf13: how about adding dates for when each of the tasks need to be done by? <a href="http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge#head-b9d1a9cb26a3e2da3d82be5fe947559bedaa745f">http://fedoraproject.org/wiki/Infrastructure/CoreExtrasMerge#head-b9d1a9cb26a3e2da3d82be5fe947559bedaa745f</a><a href="#t14:31" class="time">14:31</a>
warrenok<a href="#t14:31" class="time">14:31</a>
warrenpoelcat, well, once the trigger is pulled, almost all of it has to be done on the same day.<a href="#t14:32" class="time">14:32</a>
f13I'll email after lunch.  Headed away.<a href="#t14:33" class="time">14:33</a>
poelcatwarren: when is the trigger being pulled?<a href="#t14:34" class="time">14:34</a>
warrenpoelcat, Wednesday seems to be what was discussed today?<a href="#t14:35" class="time">14:35</a>
jeremypoelcat: that's what we were just talking about.  and it looks like wednesday is the winner<a href="#t14:35" class="time">14:35</a>
f13Ok, I'm headed out to lunch.  I'll be back in a while, send email, update wiki.<a href="#t14:36" class="time">14:36</a>
poelcatwarren: I was just thinking if we put due dates next to each of the items there would be no question of when it has to be completed by and any possible conflicts might be more apparent<a href="#t14:40" class="time">14:40</a>
poelcati'm also trying better understand the process<a href="#t14:41" class="time">14:41</a>
warrenpoelcat, agreed, except in this case almost all of it must be done on Wednesday<a href="#t14:41" class="time">14:41</a>
* poelcat would like to help with the Merge, but not sure how to best help<a href="#t14:51" class="time">14:51</a>
wwoodsI have a feeling my role will be "cheerlead, let people know what's going on, pick up slack"<a href="#t14:52" class="time">14:52</a>
wwoodssince I'm not real familiar with the build system<a href="#t14:53" class="time">14:53</a>

Generated by irclog2html.py 2.3 by Marius Gedminas - find it at mg.pov.lt!