| --- Log opened Wed Mar 04 11:03:41 2009 | ||
| * jlaska | 11:03 | |
| -!- mether [n=sundaram@nat/redhat-in/x-806c255ea5bcd1fc] has quit [Client Quit] | 11:03 | |
| wwoods | QA meeting! whee! say hi for the logs, everyone | 11:03 |
|---|---|---|
| adamw | hi | 11:04 |
| * viking_ice Greets everyone | 11:04 | |
| wwoods | As jlaska pointed out - this weekend Daylight Savings Time goes into effect in the US | 11:05 |
| * jlaska ready to save daylight! | 11:05 | |
| wwoods | the next meeting will still be at 1600UTC, but that's going to be 10AM US Eastern | 11:05 |
| * f13 | 11:06 | |
| -!- fab [n=bellet@monkey.creatis.insa-lyon.fr] has quit ["Leaving"] | 11:06 | |
| * adamw weeps | 11:06 | |
| f13 | adamw: me too | 11:06 |
| f13 | I don't know if I'll reliably make it | 11:06 |
| wwoods | yeah, it's kind of rough on the west coasters | 11:06 |
| adamw | why do we follow dst? just because utc does? it doesn't make an awful lot of sense, as people's schedules don't really change with dst. | 11:06 |
| f13 | s'ok, i'll just be semi-incoherant. | 11:07 |
| wwoods | no, we *don't* follow UTC | 11:07 |
| wwoods | errr | 11:07 |
| wwoods | we don't follow *DST* | 11:07 |
| adamw | er, yeah | 11:07 |
| wwoods | the meeting time is (currently) locked to 1600UTC | 11:07 |
| adamw | see, i'll be WORSE than this ;) | 11:07 |
| f13 | I'm awake at 7 usually due to the kid. | 11:07 |
| wwoods | so when DST changes your offset from UTC, your local meeting time changes | 11:07 |
| -!- sereinity [n=sereinit@mon69-3-82-235-39-70.fbx.proxad.net] has quit [] | 11:08 | |
| -!- JSchmitt [n=s4504kr@p4FDD1EE3.dip0.t-ipconnect.de] has joined #fedora-meeting | 11:08 | |
| wwoods | consider this another reason to petition your government to stop doing DST! | 11:08 |
| wwoods | heh | 11:08 |
| wwoods | stupid, stupid daylight savings time | 11:08 |
| jlaska | would it just be easier to change to 1700UTC? | 11:08 |
| wwoods | until that happens.. do we need to move the meeting later so we aren't screwing up the west coasters? | 11:08 |
| wwoods | right | 11:08 |
| jlaska | that itme probably works better for europe and west coast | 11:08 |
| wwoods | meeting at 1700UTC? viking_ice, does that work for you? | 11:09 |
| -!- nubs [n=anubis@fedora/nubs] has left #fedora-meeting [] | 11:09 | |
| wwoods | (any other non-North-American attendees are welcome to comment as well) | 11:09 |
| viking_ice | wwoods: Yup | 11:10 |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)] | 11:12 | |
| wwoods | the schedule has the room clear 'til 1900 | 11:12 |
| wwoods | so that works fine | 11:12 |
| jlaska | eggsellent | 11:12 |
| wwoods | okay: let's move the meeting to 1700UTC, starting next week | 11:12 |
| wwoods | which coincides with US DST, so the meeting time does *not* change for those of us following DST | 11:12 |
| jlaska | any wiki table updates needed to communicate that change? | 11:13 |
| wwoods | yup, doing that now | 11:13 |
| jlaska | thx | 11:13 |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has joined #fedora-meeting | 11:13 | |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)] | 11:13 | |
| * viking_ice sees eggsellent as some kind of American egg salad thingy | 11:13 | |
| jlaska | heh ... nah ... too much Simpsons | 11:13 |
| jlaska | although the former would be good too | 11:13 |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has joined #fedora-meeting | 11:13 | |
| wwoods | Okay, https://fedoraproject.org/wiki/Meeting_channel updated | 11:14 |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)] | 11:14 | |
| f13 | we're changing for DST, but not changing for DST (: | 11:14 |
| wwoods | I don't have an agenda, but | 11:14 |
| -!- wwoods changed the topic of #fedora-meeting to: Fedora QA | Mass Rebuild | 11:14 | |
| wwoods | yay! mass rebuild! | 11:14 |
| -!- jsmidt [n=jsmidt@vp126161.reshsg.uci.edu] has joined #fedora-meeting | 11:14 | |
| jlaska | wwoods: should we start with our action items from last week? | 11:14 |
| jlaska | just a quick run through? | 11:14 |
| wwoods | er, I suppose | 11:14 |
| -!- wwoods changed the topic of #fedora-meeting to: Fedora QA | review last week | 11:15 | |
| wwoods | http://wwoods.fedorapeople.org/fedora-qa/fedora-qa-20090225.log.html | 11:15 |
| viking_ice | depends where on the food chain you are.. animals eat grass I eat animals so I dont need to grass with my dinner | 11:15 |
| viking_ice | ken lee strikes again... | 11:15 |
| wwoods | okay, so | 11:15 |
| wwoods | - jlaska (and/or adamw) to talk to fcami and Infrastructure about the semantic mediawiki plugin | 11:15 |
| jlaska | yeah I spoke with fcami several times since that meeting | 11:16 |
| wwoods | anything to report, or just: "in progress, more details as they come in" | 11:16 |
| adamw | nothing from me, jlaska didn't ping for help | 11:16 |
| jlaska | I don't think it's the exact fit for what his focus is ... but thanks to adamw and fcami's nudging ... I'm looking into what we'd need to line up for a proof-of-concept with infrastructure | 11:16 |
| * jlaska can't type fast enough | 11:16 | |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has joined #fedora-meeting | 11:17 | |
| * jlaska finds mail link ... | 11:17 | |
| jlaska | https://www.redhat.com/archives/fedora-infrastructure-list/2009-March/msg00017.html | 11:17 |
| jlaska | so to proceed ... I'll need to get the semantic extensions packaged in fedora | 11:17 |
| -!- themayor [n=jack@ool-18b8c9c9.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)] | 11:17 | |
| jlaska | if anyone is interested in helping out on the packaging front, drop me a line please | 11:17 |
| * viking_ice points jlaska to "Typing of the dead ( http://en.wikipedia.org/wiki/The_Typing_of_the_Dead ) " should improve your typing skills | 11:18 | |
| jlaska | that's it from me on that topic ... any thoughts/concerns/suggestions? | 11:18 |
| adamw | jlaska: i can do that. | 11:18 |
| wwoods | I actually own Typing of the dead - special Dreamcast keyboard and all - so if you ever want to practice typing by KILLING ZOMBIES | 11:18 |
| wwoods | let me know | 11:18 |
| adamw | (help with packaging) | 11:18 |
| jlaska | adamw: thanks .. I'm familiarizing myself with how it's structured now ... we can chat after the meeting then? | 11:19 |
| * adamw practices typing via high-speed devel-list trolling | 11:19 | |
| adamw | jlaska: sure | 11:19 |
| jlaska | thx! | 11:19 |
| wwoods | okay, next: - autoqa "post-tree-compose" watcher and verifytree tests are top priority for autoqa (but f13 / wwoods are busy) - target completion date March 24 | 11:19 |
| wwoods | yep, mass rebuild + feature freeze + wwoods sick for a few days = no progress | 11:19 |
| f13 | yeah, no progress on this end from me | 11:19 |
| wwoods | we'll talk about that soon, once we're sure the mass rebuild / beta / etc. is sorted | 11:20 |
| wwoods | - Rawhide is still awesome | 11:20 |
| viking_ice | as always.. | 11:20 |
| wwoods | indeed. | 11:20 |
| jlaska | that's okay if no progress due to time conflicts .. .we can keep revisiting as a status checkin if we like | 11:20 |
| wwoods | - good luck to viking-ice in his archery competition | 11:20 |
| wwoods | viking_ice: how'd it go? | 11:20 |
| viking_ice | First of the Icelanders and in 13th place in my category | 11:21 |
| adamw | i remember first of icelanders wasn't very tough, right? :P | 11:21 |
| adamw | congrats | 11:21 |
| viking_ice | Since I was the only Icelander to compete.... | 11:21 |
| wwoods | heh! | 11:21 |
| wwoods | good stuff | 11:22 |
| wwoods | now if there was an *archery* of the dead game.. | 11:22 |
| wwoods | we'd be all set for training for the coming zombie apocalypse, for sure | 11:22 |
| wwoods | - wwoods to create Category:Feature_needing_QA_approval and put all unreviewed Feature pages into that category | 11:23 |
| * viking_ice notes down to create zombie foam targets | 11:23 | |
| wwoods | Done: https://fedoraproject.org/wiki/Category:Features_needing_QA_approval | 11:23 |
| wwoods | there's 34 features there (well, 33 - ignore the EmptyTemplate) | 11:23 |
| wwoods | I emailed -test-list for help reviewing | 11:24 |
| wwoods | https://www.redhat.com/archives/fedora-test-list/2009-February/msg01075.html | 11:24 |
| adamw | so, still 33 needing review? | 11:24 |
| wwoods | yes, 33 un-reviewed | 11:24 |
| f13 | isn't it a little late now that feature freeze has passed? | 11:24 |
| adamw | what's the deadline? yesterday? :) | 11:24 |
| wwoods | 2 with incomplete Scope and 1 with an incomplete test plan | 11:24 |
| f13 | or is it review to help create test cases adn such | 11:24 |
| wwoods | f13: the latter - FESCo doesn't have a problem with the contents, but we need to review the test plan and be sure we have enough info to actually test the feature | 11:25 |
| f13 | ok | 11:25 |
| wwoods | so - check the info on the feature page and ask questions to help get a plan written | 11:25 |
| adamw | ok | 11:26 |
| adamw | so, how many features *have* been reviewed? | 11:26 |
| wwoods | well, the 3 incomplete ones I mentioned, plus one approved page | 11:27 |
| wwoods | plus I did 3 last week that didn't get put into the approved category | 11:27 |
| adamw | ok, so, we're not doing well on getting people to do stuff there | 11:28 |
| viking_ice | hum we got ext4 covered already I think along with crashcatcher Fingerprint I think has an good enought test description | 11:28 |
| adamw | i'll see if i can do anything... | 11:28 |
| wwoods | okay, pages edited - there's now 4 pages marked approved | 11:28 |
| viking_ice | Empathy as well I think | 11:28 |
| wwoods | some feature pages lack the "needing approval" category | 11:29 |
| jlaska | I"m fairly comfortable about features that have had test days ... I plan to set aside some time during tomorrow's event to pick off a few | 11:29 |
| wwoods | I thought we had.. 57? something huge like that? | 11:29 |
| wwoods | anyway if you see something unmarked, feel free to review it - or just mark it as in need of review | 11:29 |
| wwoods | otherwise we're currently at 7/40 reviewed | 11:30 |
| viking_ice | hum not seeing bash 4 | 11:30 |
| wwoods | yes, please, if you think the pages we've had test days for are OK, mark them as such | 11:30 |
| jlaska | wwoods: roger, will do | 11:30 |
| viking_ice | anyway bash 4 has test cases in the source tar ball | 11:30 |
| wwoods | that's good, but.. do we have a feature page for Bash 4? | 11:32 |
| viking_ice | nope did no see it not considered as a feature? | 11:32 |
| wwoods | might be worth making a page for it | 11:32 |
| wwoods | since it has required some nontrivial integration work | 11:32 |
| wwoods | the basic test plan is pretty simple though - make sure the normal startup scripts etc. don't blow up | 11:33 |
| viking_ice | hum if bash skipped the radar I wonder what else did | 11:33 |
| wwoods | and the scope is pretty small - new bash package. | 11:33 |
| wwoods | Intel KMS wasn't a feature, originally | 11:33 |
| f13 | viking_ice: there isn't really a 'radar', there is just maintianers bringing things to the foreground | 11:34 |
| f13 | not every change requires a "feature" page | 11:34 |
| viking_ice | ok | 11:34 |
| viking_ice | anyway we can I think automate most of the tests in /bash-4.0/tests/ | 11:35 |
| wwoods | sure, if you'd like | 11:35 |
| viking_ice | more testing = good :) | 11:36 |
| wwoods | anyway, we're still reviewing last week | 11:36 |
| wwoods | - Test days for intel KMS, nouveau, AutoFontsAndMimeInstaller (dates TBD) | 11:36 |
| adamw | nouveau is scheduled and has most of the test plans up already | 11:36 |
| adamw | autofonts i didn't hear back from maintainer yet | 11:37 |
| adamw | intel KMS didn't do anything for yet, i'll try and sort that one out this week then | 11:37 |
| wwoods | intel KMS has been beaten into a fairly usable state | 11:37 |
| wwoods | so it's ready for testing | 11:37 |
| -!- themayor [n=jack@ool-18b9435e.dyn.optonline.net] has joined #fedora-meeting | 11:37 | |
| jlaska | wwoods: okay, so it didn't go as quick as I thought ... but thanks for reviewing last weeks items | 11:38 |
| viking_ice | I think we should add to this radeon and chrome testing as well and just have a big graphics test day | 11:38 |
| viking_ice | cards | 11:38 |
| adamw | viking_ice: as i said last week, i don't think that would work as well, as we'd have to have a lot of people talking about completely different things and it'd lose focus fast | 11:38 |
| -!- jnettlet [n=jnettlet@c-76-24-25-45.hsd1.ma.comcast.net] has quit [Read error: 104 (Connection reset by peer)] | 11:39 | |
| adamw | i'd like to have good testing for radeon at least, too, but probably as a separate event | 11:39 |
| adamw | didn't we just switch out the entire drm library for radeon? | 11:39 |
| wwoods | if you want to set up separate test days for radeon or openchrome, that's fine | 11:39 |
| -!- stickster is now known as stickster_afk | 11:39 | |
| jlaska | if there's any way we can also run fcami's piglit suite on each of these driver focused test days | 11:39 |
| * viking_ice is in NEEDINFO status on piglit | 11:40 | |
| adamw | if it's simple enough to install and run i could put it into the nouveau day, yes | 11:40 |
| wwoods | I thought piglit was an OpenGL test suite? | 11:40 |
| wwoods | there's no 3d support in nouveau yet, AFAIK | 11:40 |
| adamw | ah. if it's 3D, then yeah, no use. | 11:41 |
| adamw | would be most useful for the radeon stuff then. | 11:41 |
| wwoods | and intel. | 11:41 |
| jlaska | wwoods: ah yeah, you are right (3d) | 11:41 |
| jlaska | there is another upstream x test suite (non 3d iirc) ... but lemme gather some details there | 11:42 |
| wwoods | I'm gonna say | 11:42 |
| wwoods | ACTION: jlaska to investigate X test suites for intel/radeon/nouveau test days | 11:42 |
| * jlaska nods | 11:43 | |
| wwoods | okay then | 11:43 |
| -!- wwoods changed the topic of #fedora-meeting to: Fedora QA | Mass Rebuild report | 11:43 | |
| adamw | jlaska: there's one called rendercheck | 11:43 |
| wwoods | Mass rebuild is (as I understand it) complete | 11:43 |
| f13 | about 300 packages still need to be built | 11:44 |
| f13 | but that's on the maintainers | 11:44 |
| f13 | and can be done as we go | 11:44 |
| adamw | there were two major visible bugs caused by gcc 4.4 optimization issues, and some less visible ones | 11:45 |
| adamw | which makes me worry there might be regressions introduced there that we may not catch... | 11:45 |
| wwoods | the new hashes/checksums has made it so if your rpm is sufficiently old, you can't yum upgrade | 11:45 |
| wwoods | so people have been advised to 'yum upgrade rpm' first | 11:45 |
| wwoods | (side note: why don't our updater tools do that automatically?) | 11:46 |
| -!- nicubunu [n=nicubunu@mail.apsro.com] has quit [Remote closed the connection] | 11:46 | |
| f13 | wwoods: ask skvidal. Oft times new yum requires new python requires new everything including glibc, and well now what? | 11:46 |
| wwoods | adamw: right - we discussed that in the rel-eng meeting, IIRC | 11:47 |
| f13 | up2date used to do this for years and it didn't actually help | 11:47 |
| -!- warren [n=warren@redhat/wombat/warren] has joined #fedora-meeting | 11:47 | |
| wwoods | jakub is back this week and so f13 and I were going to ask if the miscompilation(s) can be identified | 11:47 |
| wwoods | so we can figure out if we need to rebuild any other packages once we have a fixed gcc | 11:47 |
| -!- giallu [n=giallu@fedora/giallu] has quit [Read error: 110 (Connection timed out)] | 11:47 | |
| wwoods | but, honestly, if something is miscompiled but it behaves correctly.. is it really miscompiled? | 11:48 |
| wwoods | so we're fixing the things that are obviously behaving wrong (vte, nspr) | 11:48 |
| adamw | wwoods: on the updater tools thing, i did make a post about that to -test-list which seems to have been roundly ignored, should i propose again on -devel-list? | 11:48 |
| adamw | or will i be set on fire again? :) | 11:48 |
| jlaska | adamw ooh, sorry I missed that one | 11:49 |
| wwoods | adamw: no, I think f13 has the essence of it | 11:49 |
| viking_ice | adamw: hum think I missed that one.. | 11:49 |
| jlaska | f13: any action required on those 300 packages prior to beta? | 11:49 |
| adamw | well, how often does python get bumped to a non-backward compatible release? | 11:49 |
| f13 | jlaska: no, they aren't beta blockers unless they are broken in some other way | 11:50 |
| wwoods | these things only happen during rawhide, and, well, that's the sort of workaround that's required to run rawhide | 11:50 |
| f13 | adamw: every couple releases. | 11:50 |
| jlaska | f13: gotcha | 11:50 |
| adamw | f13: right, not very often. well, anyhow. could be a long discussion | 11:50 |
| f13 | adamw: that's once maybe twice a year | 11:50 |
| wwoods | it would be better if we had a better way of getting this information to rawhiders | 11:50 |
| adamw | i should point out, though, that it helps with another situation, which i described in the list post | 11:50 |
| f13 | how often do you have a real situation where updating the updater first actually fixes anything? | 11:50 |
| adamw | f13: in mdv i can think of, oh, two or three. they were quite big ones, though. :) and it's not the *same* time as python gets updated to a new release... | 11:51 |
| wwoods | but, basically, it sounds like the added complexity doesn't actually solve most of the observed problems | 11:51 |
| f13 | adamw: I guess we're just more careful with our update tools (: | 11:51 |
| wwoods | and it doesn't affect proper major-version-to-major-version upgrades | 11:51 |
| jlaska | up2date had the update the updater concept ... and it never worked | 11:51 |
| adamw | wwoods: it would've solved this one. that's not the objection - the objection is that yum's dep chain is sufficiently long that it has no measurable effect. doesn't answer the question of just doing it for rpm, though. | 11:52 |
| jlaska | well sorry, it worked ... but it's very flaky | 11:52 |
| adamw | wasn't that just because up2date was flaky? :P | 11:52 |
| jlaska | heh, part of it :D | 11:52 |
| wwoods | the objection seems more general - the added complexity only fixes a very small class of problems | 11:52 |
| jlaska | it never forked itself properly (don't quote me on that please) :) | 11:53 |
| f13 | adamw: if you're hitting a case wher eyou need a newer rpm to finish the trasaction, you often need a newer yum too to work with that newer rpm, and you're back to square one | 11:53 |
| wwoods | and there's a much larger set of similar problems with similar solutions | 11:53 |
| wwoods | i.e. "update this other thing manually first" | 11:53 |
| adamw | often, yes. not always. well, i won't push it. i just don't like the "only a few" fallacy | 11:53 |
| adamw | any given fix usually only fixes a few issues | 11:53 |
| adamw | if there were lots of cases where we could do one fix and fix a giant bunch of problems, we could finish fedora for ever next week and take some time off :P | 11:54 |
| f13 | adamw: I and others just don't see the added complexity and flakyness as worth trying to solve the rare cases where it would help | 11:54 |
| f13 | I'd rather just avoid those cases, particularly in a stable release | 11:54 |
| adamw | fair enough. i reserve the right to point and laugh when the situation i outlined in my email happens, though. :) | 11:55 |
| wwoods | anyway | 11:55 |
| wwoods | sorry, that was a tangent | 11:55 |
| adamw | yeah, my fault. sorry | 11:55 |
| wwoods | (but a rawhide-user-notifier program is something I've had kicking around my head for a while - should reinvestigate that) | 11:56 |
| wwoods | so, yes, we're going to try to figure out if we can automatically identify and rebuild other miscompiled builds | 11:57 |
| wwoods | failing that, we'll fix the bugs as they come up | 11:57 |
| wwoods | which has worked fine so far | 11:57 |
| wwoods | still about 300 packages left unbuilt, as f13 mentioned | 11:57 |
| wwoods | how are we doing with respect to rawhide composes, media composes, and live images? | 11:58 |
| wwoods | last I heard, we couldn't build live images due to.. syslinux problems? | 11:58 |
| f13 | not so good | 11:58 |
| jlaska | missing i386/os/pxeboot/vmlinuz today I think ... x86_64 looks good today | 11:58 |
| f13 | we're getting images on i386 and x86_64 with patches that are upstream in anaconda | 11:58 |
| f13 | but those images don't actually make it past init yet | 11:58 |
| f13 | and I'm really at a loss as to why | 11:58 |
| f13 | and efi images are kind of screwed up | 11:59 |
| f13 | the entire anaconda team is wrapped up in storage rewrite and efi features | 11:59 |
| jlaska | f13: are these official rawhide images, or private test images you are constructing? | 11:59 |
| f13 | private test images, there hasn't been a build of anaconda with those chnages yet | 11:59 |
| jlaska | gotcha | 11:59 |
| f13 | well there was one yesterday | 12:00 |
| f13 | but I think its missing lot sof stuff | 12:00 |
| f13 | jeremy threw a patch to fix livecd for syslinux yesterday | 12:00 |
| f13 | but he mentioned other issues he was working on. | 12:00 |
| f13 | so no new build there either | 12:00 |
| -!- j6k [n=j6k@195-240-118-19.ip.telfort.nl] has joined #fedora-meeting | 12:01 | |
| wwoods | so we're currently unable to build boot images or build media, and we can't update from F11Alpha to current rawhide | 12:01 |
| f13 | why can't you update? | 12:02 |
| wwoods | can we pick a pre-Rebuild rawhide to use as our suggested install base? | 12:02 |
| wwoods | well, not without manually updating RPM first | 12:02 |
| wwoods | it's just a bit tricky | 12:02 |
| f13 | it's two commands instead of one | 12:02 |
| wwoods | but we've got some known-good rawhide sitting around | 12:03 |
| f13 | oooooooh tricky | 12:03 |
| f13 | (: | 12:03 |
| jlaska | I'm discussing which rawhide to use for tomorrow's test day with dlehman ... we'll likely be using an updates.img in conjunction with a previously working build | 12:03 |
| adamw | how is this going to affect thursday's test day? does a network install of rawhide work? | 12:03 |
| wwoods | I know I know, but if there's a known-good rawhide that doesn't need that | 12:03 |
| jlaska | I'm not clear if they'll have all their commits and and a new anaconda rebuilt (that works) for tomorrow | 12:04 |
| -!- nphilipp [n=nils@nat/redhat/x-dc324bae42407214] has quit ["Leaving"] | 12:04 | |
| jlaska | that's a lot to line up | 12:04 |
| jlaska | but heck if it's there we'll definitely use it | 12:04 |
| -!- drago01 [n=linux@chello062178124130.3.13.univie.teleweb.at] has joined #fedora-meeting | 12:04 | |
| wwoods | jlaska: cool - if you settle on a stable one, please send a note to -test-list? | 12:05 |
| wwoods | otherwise we'll just continue making people deal with the horribly complex hack of typing 'yum upgrade rpm' first | 12:05 |
| wwoods | heh | 12:05 |
| jlaska | yeah it's really up to what works for the anaconda-devel folks | 12:05 |
| jlaska | I must have been lucky ... I didn't explicitly do 'yum upgrade rpm' | 12:06 |
| -!- drago01 [n=linux@chello062178124130.3.13.univie.teleweb.at] has quit [Remote closed the connection] | 12:06 | |
| jlaska | or I have a nasty bug lurking? | 12:06 |
| wwoods | jlaska: right, it only happens if you're going from a sufficiently old RPM to the Post-Rebuild packages | 12:06 |
| wwoods | (if I understand right) | 12:06 |
| jlaska | ah gotcha | 12:06 |
| adamw | yeah. | 12:06 |
| adamw | if you were updating daily you don't hit it. | 12:06 |
| wwoods | if you updated after F11Alpha, you're fine | 12:06 |
| jlaska | f13: do you have time after the meeting to chat in #anaconda about tomorrows rawhide build? | 12:07 |
| wwoods | but going straight from F11Alpha to the Post-Rebuild package set will blow up | 12:07 |
| f13 | jlaska: after I fetch some breakfast I do | 12:07 |
| jlaska | f13: okay ... thanks. I think we need to coordinate a few things | 12:07 |
| wwoods | also, obviously, yum upgrades from F10 to Rawhide will break in the same way | 12:08 |
| f13 | actually no | 12:08 |
| wwoods | really? | 12:08 |
| f13 | the latest F10 rpm supports the larger hashes | 12:08 |
| nirik | (if they didn't update rpm in f10 to the latest update version first) | 12:08 |
| wwoods | did we push an RPM update? oh nice | 12:08 |
| f13 | but you still have to yum update rpm, then yum update | 12:08 |
| f13 | as in get the latest f10 rpm, /then/ leap to rawhide | 12:09 |
| wwoods | gotcha | 12:09 |
| f13 | GOLD F10 to rawhide won't work. | 12:09 |
| adamw | it would if yum updated rpm first! ;) (/me ducks and runs) | 12:09 |
| adamw | but i don't see that causing any major issues in practice | 12:09 |
| -!- fab [n=bellet@bellet.info] has joined #fedora-meeting | 12:10 | |
| adamw | why would you do a yum upgrade on a gold f10, in practice | 12:10 |
| wwoods | ah okay. so "sufficiently old RPM" includes F11Alpha and F10 Gold, but not F10+updates | 12:10 |
| wwoods | adamw: people do all kinds of crazy shit | 12:10 |
| adamw | true. what i mean is, we can tell 'em they're being crazy | 12:11 |
| wwoods | every time a developer asks "why would anyone ever do that?" someone, somewhere *immediately* starts doing exactly that | 12:11 |
| f13 | adamw: I think F10 -> rawhide rpm would require some glibc bits along the way | 12:11 |
| -!- abadger19991 [n=abadger1@65.78.187.8] has joined #fedora-meeting | 12:12 | |
| jlaska | should we move this discussion to #fedora-qa after the meeting ... or continue here? | 12:12 |
| wwoods | sorry, yeah | 12:12 |
| -!- abadger1999 [n=abadger1@65.78.187.8] has quit [Nick collision from services.] | 12:13 | |
| -!- abadger19991 is now known as abadger1999 | 12:13 | |
| wwoods | so we should be lending a hand to the anaconda guys if possible | 12:13 |
| wwoods | but currently we're not able to build anything installable | 12:13 |
| wwoods | hopefully that will be fixed RSN. | 12:13 |
| jlaska | I'm able to install with last nights x86_64 images ... but I don't think that has the anaconda storage changes yet (iirc) | 12:14 |
| wwoods | oh it's just i386? I thought we weren't building any images | 12:14 |
| jlaska | I'm not clear on that ... I don't see a pxeboot/vmlinuz for i386 from last night ... but x86_64 has all the images I needed for install | 12:14 |
| f13 | jlaska: no, storage changes are still on a branch. | 12:15 |
| jlaska | anyway, we can chat about this over in #anaconda once f13 and anaconda-devel folks have ingested some calories | 12:15 |
| wwoods | right | 12:15 |
| wwoods | so we *are* building rawhide images, live composes are actively being worked on | 12:16 |
| wwoods | media composes.. unknown | 12:16 |
| wwoods | I'm guessing it'll all be sorted by next meeting | 12:16 |
| wwoods | anything else we should discuss? | 12:17 |
| -!- wwoods changed the topic of #fedora-meeting to: Fedora QA | misc | 12:17 | |
| -!- mcepl [n=mcepl@49-117-207-85.strcechy.adsl-llu.static.bluetone.cz] has left #fedora-meeting [] | 12:17 | |
| -!- tatica [n=tatica@nelug/designer/tatica] has quit [Remote closed the connection] | 12:18 | |
| -!- jsmidt [n=jsmidt@vp126161.reshsg.uci.edu] has quit [Success] | 12:19 | |
| -!- tatica [n=tatica@nelug/designer/tatica] has joined #fedora-meeting | 12:19 | |
| * wwoods takes that as a no | 12:19 | |
| -!- buggbot [n=supybot@landfill.bugzilla.org] has quit [Read error: 104 (Connection reset by peer)] | 12:19 | |
| jlaska | any updates on traige activities? | 12:19 |
| jlaska | sorry for delay | 12:19 |
| jlaska | triage even | 12:19 |
| adamw | well, we have the bugzappers meeting for that :) but, yeah, it's pretty active lately | 12:20 |
| adamw | we had quite a lot of new people last week, changes for the wiki are being worked on, first triage day was yesterday | 12:20 |
| viking_ice | there seems to be a bunch of new bugzappers joining | 12:20 |
| jlaska | right ... just curiuos about any high-level take aways | 12:20 |
| viking_ice | did you offer free beer somewhere | 12:20 |
| -!- buggbot [n=supybot@landfill.bugzilla.org] has joined #fedora-meeting | 12:21 | |
| jlaska | yeah I'vebeen blown away with the introductions you've been getting ... nice work to the bugzappers | 12:21 |
| adamw | viking_ice: they're all me, using gmail aliases and fake moustaches | 12:22 |
| adamw | ;) | 12:23 |
| jlaska | doesn't matter ... if you can sustain it, rock on ;) | 12:23 |
| adamw | but no, actually, i think things are moving from qa to bugzappers rather than the other way around - i think some of the new interest is coming via the updated QA wiki pages for e.g. | 12:23 |
| -!- loupgaroublond [n=loupgaro@82-171-65-13.ip.telfort.nl] has joined #fedora-meeting | 12:23 | |
| adamw | for now we're sort of scrambling to keep up with a process for getting the new people started | 12:23 |
| jlaska | so the cleaned up QA pages are properly routing folks to sub-teams | 12:23 |
| jlaska | cool | 12:23 |
| adamw | so i guess the takeaway might be, make sure we have stuff for new people to do immediately | 12:24 |
| viking_ice | nice | 12:24 |
| -!- fbijlsma [n=fbijlsma@217.110.39.107] has joined #fedora-meeting | 12:24 | |
| -!- fbijlsma [n=fbijlsma@217.110.39.107] has quit [Read error: 104 (Connection reset by peer)] | 12:24 | |
| viking_ice | get them attache to one component and master it ( join upstream mailing lists etc.. ) | 12:24 |
| viking_ice | then they can monitor that one to begin with then add on as the experience grows.. | 12:25 |
| jlaska | adamw: good observation ... we're struggling a bit with defining a play area for beaker contributors. I've got a discussion coming up later on to try and outline some things to hopefully improve that | 12:25 |
| viking_ice | probably the way I suggest we go with new testers | 12:25 |
| jlaska | wwoods: as for action items for the week ... I only had a few ... unless I missed stuff | 12:26 |
| jlaska | * [jlaska+adamw] - update on semantic packaging | 12:26 |
| jlaska | * [jlaska] - review upstream xorg test suite status (in addition to piglit) ... for use during upcoming Xorg driver specific test day | 12:26 |
| jlaska | * [wwoods] - discuss with f13 + jakub a method for identifying similar gcc4.4 optimization bugs? | 12:26 |
| jlaska | is that accurate ... what did I miss? | 12:26 |
| -!- itamarjp [n=itamar@187-032-068-225.static.ctbctelecom.com.br] has quit [Read error: 145 (Connection timed out)] | 12:28 | |
| wwoods | jlaska: email fedora-test-list if (when) you identify a known-good rawhide tree? | 12:28 |
| jlaska | okay | 12:29 |
| * jlaska jumps for meeting conflict | 12:30 | |
| adamw | i guess i should talk to fcami about maybe doing a radeon test day | 12:30 |
| adamw | sound good? | 12:30 |
| f13 | f13 + jlaska to work with anaconda team to identify a rawhide tree for test day | 12:30 |
| jlaska | ooh yes | 12:30 |
| adamw | and intel kms | 12:30 |
| wwoods | radeon test day, intel test day | 12:30 |
| adamw | OK | 12:30 |
| adamw | they're on my todo list then | 12:30 |
| wwoods | oh, and: [all] - review feature pages in https://fedoraproject.org/wiki/Category:Features_needing_QA_approval | 12:32 |
| wwoods | (esp. features that have had test days, which can probably move straight to Category:QA approved feature pages) | 12:32 |
| -!- mbacovsk [n=mbacovsk@nat/redhat/x-45b7b07583137973] has quit [Read error: 113 (No route to host)] | 12:32 | |
| wwoods | I think that's it, and we're over time | 12:33 |
| wwoods | any further discussion can go to #fedora-qa. thanks for your time, folks. | 12:33 |
| --- Log closed Wed Mar 04 12:33:38 2009 | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!