| === doko__ is now known as doko | ||
| stgraber | infinity: are you still around for another 30min or so? I may need you to change ~/.isotracker.conf on nusakan | 02:57 |
|---|---|---|
| stgraber | infinity: http://iso.qa.ubuntu.com is almost good to go, so will need to switch the target once I kill http://91.189.93.73 (so we avoid daily builds showing up on the wrong one) | 02:58 |
| stgraber | slangasek: ^ | 03:08 |
| stgraber | we're ready for the switch now | 03:08 |
| stgraber | just replace http://91.189.93.73 to http://iso.qa.ubuntu.com in ~/.isotracker.conf (cdimage user I believe) | 03:08 |
| stgraber | morning pitti | 05:10 |
| stgraber | pitti: can you update ~/.isotracker.conf (see above)? | 05:11 |
| pitti | good morning | 05:16 |
| pitti | stgraber: done | 05:18 |
| pitti | stgraber: but it says "still under maintenance", is that expected? | 05:18 |
| stgraber | pitti: yes, I was waiting for someone to switch the target of the dailies before I switch it to production mode and disable the temporary one | 05:19 |
| pitti | ah :) | 05:20 |
| stgraber | ok, opened the new one and closed 91.189.93.73. We now have an all new and shiny tracker at http://iso.qa.ubuntu.com | 05:21 |
| stgraber | Seems like everything is working, so calling it a day, back in 8 hours or so :) | 05:22 |
| pitti | stgraber: still says "under maintenance" | 05:23 |
| pitti | stgraber: thanks, sleep well! | 05:23 |
| stgraber | pitti: argh, that's annoying... let me hit the cache harder | 05:24 |
| stgraber | pitti: any better now? | 05:24 |
| pitti | ah, yes! | 05:25 |
| stgraber | Drupal's aggressive caching is really, well, aggressive :) | 05:25 |
| Laney | Somebody points out in #-motu that DebianImportFreeze mentions unstable, and that the release schedule links there instead of LTSDebianImportFreeze. Which fix should we do? | 10:05 |
| infinity | Laney: Documenting DIF twice seems odd. Combining the two pages into one that explains why it's sometimes unstable and sometimes testing might make more sense. | 11:12 |
| Laney | I thought it would be easiest to just remove 'unstable' | 11:12 |
| infinity | Laney: (But for now, linking the PreciseReleaseSchedule to LTSDIF works, I guess) | 11:12 |
| Laney | it's not as if that document is the canonical reference for the dafault sync source | 11:12 |
| infinity | Well, both pages seem subtly different too. | 11:13 |
| Laney | LTS... says "(and tested)" and does not mention autosyncs of NEW packages. | 11:26 |
| Laney | I do not think it's true that new packages aren't autosynced from testing. | 11:26 |
| infinity | It's not true, no. We still autosync. | 11:27 |
| Laney | so AFAICS the differences aren't meaningful | 11:28 |
| infinity | I wouldn't say so, no. | 11:28 |
| Laney | we can check with skaet later I guess | 11:28 |
| infinity | I think instead of saying "from Debian <release>", another line that explains that sometimes we sync from unstable and sometimes from testing, and this is determined at the beginning of each cycle by the release team... | 11:29 |
| infinity | (While we more or less have followed the non-LTS->unstable, LTS->testing trend, I see no reason to state that as a future fact) | 11:30 |
| Laney | indeed, it should be a per-release decision | 11:31 |
| Laney | there should be a place to document it | 11:31 |
| infinity | Well, LP knows the parent series. | 11:32 |
| infinity | At least, it does now. :P | 11:32 |
| Laney | ah, yes, lp/ubuntu/release | 11:32 |
| infinity | I think documenting these things in too many places is just asking for them to be out of sync. | 11:33 |
| infinity | And LP is authoritative on the matter, now that syncing is a native affair. | 11:33 |
| Laney | There should be one place that things point to. That's what I was after. Even better if it's used by tools too. | 11:34 |
| ogra_ | doe that look ok for adding a "manual build" notification to the failure mails ? http://paste.ubuntu.com/764832/ | 11:57 |
| ogra_ | *does | 11:57 |
| ogra_ | (in cdimage that is) | 11:58 |
| * ogra_ would like a typo check before committing ... i cant see any issue but that doesnt mean much :P | 11:59 | |
| infinity | ogra_: I'd probably write it as MANUALTEXT="(built by ${SUDOUSER}) " | 12:02 |
| ogra_ | k, no prob | 12:03 |
| infinity | ogra_: And smoosh it right against $PROJECT (otherwise, you get an extra space when it's not set) | 12:03 |
| ogra_ | yeah, well, whats an extra space :P | 12:03 |
| infinity | s/SUDOUSER/SUDO_USER/ | 12:03 |
| infinity | Extra spaces are the devil's work! | 12:03 |
| ogra_ | heh | 12:04 |
| infinity | ogra_: Need a trailing space. | 12:05 |
| infinity | MANUTEXT="(built by ${SUDO_USER}) " | 12:05 |
| ogra_ | === modified file 'bin/buildlive' | 12:05 |
| ogra_ | --- bin/buildlive2011-12-02 06:14:28 +0000 | 12:05 |
| ogra_ | +++ bin/buildlive2011-12-09 12:04:52 +0000 | 12:05 |
| ogra_ | @@ -20,6 +20,11 @@ | 12:05 |
| ogra_ | DIST="$3" | 12:05 |
| ogra_ | fi | 12:05 |
| ogra_ | 12:05 | |
| ogra_ | +MANUTEXT= | 12:05 |
| ogra_ | +if [ -n "${SUDO_USER}" ]; then | 12:05 |
| ogra_ | +MANUTEXT="(built by ${SUDO_USER})" | 12:05 |
| ogra_ | +fi | 12:05 |
| ogra_ | + | 12:05 |
| ogra_ | datestamp="$(date +%Y%m%d)" | 12:05 |
| ogra_ | 12:05 | |
| infinity | pasting here might not have been clever. :P | 12:05 |
| ogra_ | RES="" | 12:05 |
| ogra_ | @@ -118,7 +123,7 @@ | 12:05 |
| ogra_ | ADDRESSES="$(get_notify_addresses "${PROJECT%-dvd}")" | 12:06 |
| ogra_ | if [ "$ADDRESSES" ]; then | 12:06 |
| ogra_ | wget -q -O - "http://$machine/~buildd/LiveCD/$DIST/$PROJECT${SUBPROJECT:+-$SUBPROJECT}${subarch:+-$subarch}${UBUNTU_DEFAULTS_LOCALE:+-$UBUNTU_DEFAULTS_LOCALE}/latest/livecd-${datestamp}-${arch}.out" | \ | 12:06 |
| ogra_ | - mail -s "LiveFS $PROJECT${SUBPROJECT:+-$SUBPROJECT}${subarch:+-$subarch}${UBUNTU_DEFAULTS_LOCALE:+-$UBUNTU_DEFAULTS_LOCALE}/$DIST/$arch failed to build on $datestamp" \ | 12:06 |
| ogra_ | + mail -s "LiveFS $MANUTEXT$PROJECT${SUBPROJECT:+-$SUBPROJECT}${subarch:+-$subarch}${UBUNTU_DEFAULTS_LOCALE:+-$UBUNTU_DEFAULTS_LOCALE}/$DIST/$arch failed to build on $datestamp" \ | 12:06 |
| ogra_ | -a 'X-Generated-By: buildlive' \ | 12:06 |
| ogra_ | $ADDRESSES | 12:06 |
| ogra_ | fi | 12:06 |
| ogra_ | argh ! | 12:06 |
| ogra_ | sorry | 12:06 |
| * ogra_ czrses to not have middle click ... | 12:06 | |
| ogra_ | http://paste.ubuntu.com/764840/ was what i meant to paste | 12:06 |
| infinity | ogra_: Also, I'm reviewing on nusakan, no need to paste at all. :P | 12:06 |
| ogra_ | haha | 12:06 |
| ogra_ | k | 12:06 |
| ogra_ | well, i'll merge if you dont scream now | 12:06 |
| infinity | ogra_: s/${SUDO_USER})"/${SUDO_USER}) "/ | 12:06 |
| ogra_ | fixed | 12:07 |
| infinity | Lovely. | 12:07 |
| infinity | The name of the variable is a bit odd. | 12:07 |
| ogra_ | eek | 12:08 |
| ogra_ | ogra@nusakan:/srv/cdimage.ubuntu.com$ bzr pull | 12:08 |
| ogra_ | Using saved parent location: /srv/cdimage.ubuntu.com/bzr/private/cdimage | 12:08 |
| ogra_ | M bin/buildlive | 12:08 |
| ogra_ | Text conflict in bin/buildlive | 12:08 |
| ogra_ | 1 conflicts encountered. | 12:08 |
| ogra_ | hmpf | 12:08 |
| ogra_ | i'm sure my tree in ~ is up to date | 12:08 |
| ogra_ | hmm, someone seems to commit directly to the production tree | 12:09 |
| ogra_ | grmbl | 12:10 |
| infinity | Or not commit. | 12:10 |
| infinity | It was uncommitted changes you ran into. | 12:10 |
| ogra_ | oh | 12:10 |
| infinity | And wrong ones, at that... | 12:11 |
| ogra_ | oh my | 12:11 |
| infinity | Is that Michael doing weird things to make jani's live build go? :/ | 12:11 |
| ogra_ | he shouldnt have to change anything | 12:11 |
| ogra_ | NCommander, ^^^^ ???? | 12:11 |
| ogra_ | he only needs to set a var to not make it publish when running the build | 12:12 |
| infinity | Yeah, but that s/arch/cpuarch/ thing looks pretty suspect. | 12:12 |
| infinity | Oh, wait, that's just log fetching? | 12:12 |
| infinity | Whatever. | 12:12 |
| infinity | Just manually merge your bits. | 12:12 |
| infinity | Or I will. | 12:13 |
| ogra_ | feel free | 12:13 |
| infinity | There. Merged. | 12:13 |
| * ogra_ looks at default-arches now to enable hf for desktop and server | 12:13 | |
| infinity | And whoever's uncommited crap that is better claim it. | 12:13 |
| ogra_ | i'll run into the same issue with the next commit them i guess | 12:14 |
| ogra_ | *then | 12:14 |
| infinity | Only if you commit to buildlive again... | 12:14 |
| infinity | In those same lines. :P | 12:14 |
| ogra_ | no, to etc/default-arches only | 12:14 |
| infinity | Yeah, default-arches has no uncommitted changes. | 12:14 |
| infinity | You're fine. | 12:14 |
| ogra_ | i see far more issues than buildlive here | 12:14 |
| ogra_ | etc/purge-days | 12:15 |
| ogra_ | etc/config | 12:15 |
| infinity | Yeah, I know. There are lots of bits uncommitted. | 12:15 |
| infinity | But not the file you're about to edit. | 12:15 |
| ogra_ | bin/publish-release | 12:15 |
| infinity | (No need to paste this all) | 12:15 |
| ogra_ | k, i trust you :) | 12:15 |
| ogra_ | infinity, can you take a look ? i'm not sure about the server bit ... | 12:17 |
| infinity | ogra_: Eh? | 12:18 |
| === ara is now known as Guest31779 | ||
| ogra_ | the change to default arches | 12:18 |
| infinity | ogra_: Yeah, that was wrong. :) | 12:19 |
| infinity | I'll fix. | 12:19 |
| ogra_ | heh | 12:19 |
| ogra_ | i'll never understand that thing | 12:20 |
| ogra_ | sigh | 12:20 |
| infinity | Oh, actually. It might have been accidentally right. | 12:21 |
| infinity | ubuntu-server/preinstalled matches twice. | 12:21 |
| ogra_ | only for maverick builds, no ? | 12:21 |
| ogra_ | oh, its maverick- | 12:21 |
| ogra_ | ignore me :P | 12:21 |
| infinity | Ahh, no. I was right. | 12:23 |
| ogra_ | the bottom thing needs to go and the maverick- line needs hf, right ? | 12:23 |
| infinity | Yeah. | 12:24 |
| ogra_ | s/to go/reverting/ | 12:24 |
| ogra_ | k, then i understand it right | 12:24 |
| infinity | Already fixing. | 12:24 |
| ogra_ | (until i want to touch it again and forgot all tis next time :P ) | 12:24 |
| infinity | Meh. I really should move all this armhf stuff to precise- lines. | 12:26 |
| infinity | Incoming mess. | 12:26 |
| ogra_ | that would at least make it easier to understand | 12:26 |
| ogra_ | i really struggel with all these wildcards ... especially if i havent touched it fro a few months | 12:27 |
| infinity | Trust me, I didn't just make it easier to read. :P | 12:27 |
| ogra_ | (and i need typing training) | 12:27 |
| infinity | adconrad@nusakan:~/cdimage$ CDIMAGE_ROOT=. ALL_DISTS="lucid maverick natty oneiric precise" bin/default-arches ubuntu-server daily-preinstalled oneiric | 12:28 |
| infinity | armel+omap armel+omap4 | 12:28 |
| infinity | adconrad@nusakan:~/cdimage$ CDIMAGE_ROOT=. ALL_DISTS="lucid maverick natty oneiric precise" bin/default-arches ubuntu-server daily-preinstalled precise | 12:28 |
| infinity | armel+omap armel+omap4 armhf+omap armhf+omap4 | 12:28 |
| infinity | Better. | 12:28 |
| ogra_ | as elegant it is, i must admit i found the cluttered crontab better | 12:28 |
| * infinity commits. | 12:28 | |
| infinity | ogra_: Fixed in production now. | 12:29 |
| ogra_ | yep, just pulled | 12:29 |
| infinity | ogra_: Calling default-arches helps to figure out what you've done. | 12:29 |
| ogra_ | ah, cool, i didnt even know there was a binary | 12:31 |
| ogra_ | (i could indeed have looked ... but i have a slighly distracted day today) | 12:31 |
| infinity | It's not the most intuitive thing to call, mind you. | 12:31 |
| infinity | But at least you can do it from your branch. | 12:31 |
| ogra_ | yeah, seeing that above | 12:31 |
| infinity | (See above for calling convention) | 12:31 |
| infinity | I'm going to catch a nap before morning meetings and work. | 12:32 |
| ogra_ | can you do the release meeting ? | 12:33 |
| ogra_ | since michael doesnt seem able to | 12:33 |
| infinity | I have a conflicting linaro meeting. :/ | 12:33 |
| ogra_ | ah, k, i'll try to get tobin to do it then | 12:33 |
| infinity | I can fix that, but not for the one in a few hours. | 12:33 |
| ogra_ | yeah | 12:34 |
| ogra_ | its only for today anyway ... | 12:34 |
| stgraber | pitti: didn't we get Ubuntu dailies today? | 14:03 |
| pitti | stgraber: we did | 14:04 |
| pitti | http://cdimage.ubuntu.com/daily-live/20111209.1/ | 14:04 |
| pitti | http://cdimage.ubuntu.com/daily/20111209/ | 14:04 |
| stgraber | hmm, ok, they aren't on the tracker... | 14:04 |
| pitti | $ cat .isotracker.conf | 14:04 |
| pitti | [general] | 14:04 |
| pitti | url=http://iso.qa.ubuntu.com/xmlrpc.php | 14:04 |
| pitti | username=ubuntu-cdimage | 14:04 |
| pitti | password=<secret> | 14:05 |
| pitti | default_milestone=Precise Daily | 14:05 |
| pitti | stgraber: looks ok? | 14:05 |
| stgraber | yep, that looks good | 14:05 |
| stgraber | checking the build logs now, maybe we'll get a nice python trace in there :) | 14:05 |
| stgraber | ok, I see a "Failed to add build to the tracker" | 14:07 |
| stgraber | now to figure out why exactly | 14:07 |
| stgraber | ok, got the same from here at least | 14:08 |
| ogra_ | i might be late to the release meeting ... | 14:08 |
| * ogra_ needs to pick up his car from the other side of the city and the shop closes right after the meeting | 14:09 | |
| ogra_ | (please someone forward that to kate if i dont manage) | 14:09 |
| stgraber | pitti: digging a bit more, the API seems to work, though not the admin API for some reason. While I do that, do you have a way to get a list of everything that build since "Ubuntu Server i386" (20111209)? | 14:15 |
| pitti | stgraber: we didn't do any manual builds today, so we should have exactly one build for each flavour | 14:19 |
| pitti | i. e. just the cronjobs | 14:19 |
| pitti | server starts at 06:29 | 14:20 |
| pitti | so that would be: ubuntu desktop/alternate | 14:20 |
| ogra_ | note that we autobuild armhf images from today on ... | 14:20 |
| pitti | kubuntu desktop/alternate | 14:20 |
| * ogra_ hopes that doesnt confuse anything on the tracker | 14:20 | |
| pitti | edubuntu dvd | 14:20 |
| pitti | kubuntu dvd | 14:20 |
| pitti | sorry, scratch edubuntu, that goes earlier | 14:21 |
| pitti | studio, mythbuntu, lubutun | 14:21 |
| pitti | that's it | 14:21 |
| pitti | everything else builds before server | 14:21 |
| pitti | stgraber: ^ | 14:21 |
| stgraber | pitti: ok, thanks. I found the problem (squid server in front of the tracker blocks http auth), so only the public API works. I'm talking to IS to have that fixed, then I'll just push all of these again | 14:23 |
| stgraber | ogra_: I don't think we have armhf on the tracker and probably not in our mapping table so it shouldn't be a problem, though we'll need to add it | 14:35 |
| stgraber | pitti: I think I fixed them all by hand, in the process I noticed a few don't seem to get posted automatically, likely because of a wrong or missing mapping in the script. That's: Kubuntu Mobile armel+omap3, Kubuntu Mobile i386, Wubi i386 and Wubi amd64 | 14:40 |
| stgraber | (these all had daily builds today but never had a daily registered on the tracker) | 14:40 |
| stgraber | skaet: good morning | 14:45 |
| stgraber | skaet: and look at our shiny new http://iso.qa.ubuntu.com :) | 14:45 |
| stgraber | (still talking to IS about the authenticated API currently not working) | 14:45 |
| skaet | stgraber, *\o/* | 14:46 |
| pitti | skaet: good morning | 14:46 |
| pitti | skaet: FYI, kenvandine will join the release meeting today instead of me, as I need to leave a little earlier | 14:46 |
| skaet | pitti, good afternoon. :) | 14:46 |
| skaet | pitti, ok. | 14:47 |
| skaet | thanks for letting me know. | 14:47 |
| skaet | stgraber, delighted to see the new tracker - awesome way to start the day. | 14:48 |
| === ara_ is now known as ara | ||
| gilir | hi, all lubuntu desktop daily isos are currently not published, I think there is a little problem in the publisher script : http://paste.ubuntu.com/765014/ | 15:37 |
| stgraber | pitti: add lubuntu to the list I gave you before (of things building with a wrong/missing mapping) :) | 15:39 |
| Laney | skaet: earlier we were talking about merging LTSDebianImportFreeze with DebianImportFreeze — noticed because the latter is linked from the release schedule and still says that we are syncing from unstable | 15:40 |
| Laney | what do you think? | 15:40 |
| stgraber | gilir: is that actually making the build fail or just not publish them on the tracker? | 15:40 |
| stgraber | gilir: because when I added them all manually this morning (problem with the API...) I didn't see a 20111209 build of lubuntu | 15:41 |
| gilir | stgraber, I think only the publishing, I didn't see any failure in the log about the build | 15:41 |
| stgraber | gilir: ah right, just saw the timestamp, that was yesterday's. So the tracker is up to date | 15:43 |
| skaet | Laney, oversite, and yes will fix. | 15:46 |
| skaet | :) | 15:46 |
| skaet | thanks for flagging. | 15:46 |
| Laney | AFAICS we should remove the word "unstable" from DebianImportFreeze and add a sentence saying to look at launchpad.net/ubuntu/<release> to see where we are syncing from | 15:48 |
| * ogra_ is back ... just in time it seems :) | 15:53 | |
| skaet | Laney that's a better solution, agreed. | 15:57 |
| Laney | :-) | 15:58 |
| * NCommander waves (sorry for being a bit tardy) | 16:00 | |
| ogra_ | NCommander, i'm here, no need to cover ... | 16:00 |
| ogra_ | i just made it in time | 16:00 |
| Laney | looks like current ben is still compatible with the old input format | 16:57 |
| Laney | and it also now contains code to generate an index | 16:57 |
| Laney | win and win | 16:57 |
| scott-work | skaet: sorry i missed the meeting, work was inescapable this morning | 19:23 |
| skaet | scott-work, no worries. | 19:24 |
| skaet | scott-work, I'll be posting links to all the other teams, and the IRC minutes in the agenda today. | 19:25 |
| skaet | Laney, https://wiki.ubuntu.com/DebianImportFreeze | 21:06 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!