astraljava | cjwatson: Yay! Thanks! | 06:20 |
---|---|---|
jamespage | good morning - please could an archive admin accept the libspring-java binaries packages in the NEW queue for oneiric - ta | 08:05 |
=== jibel__ is now known as jibel | ||
=== doko__ is now known as doko | ||
Daviey | apw: Confirming that the PPA kernel resolves the issue | 10:08 |
apw | skaet, cjwatson, i assume that the PPA kernel should be copied out to -proposed at least, also have we decided on how much QA to apply | 10:39 |
cjwatson | jibel: any showstoppers you know of that would prevent releasing non-server images? | 10:41 |
jibel | cjwatson, I found a problem with ltsp install | 10:41 |
jibel | it tried to download the files from the wrong location on the ltsp server | 10:42 |
jibel | but there's an easy workaround | 10:42 |
cjwatson | is it a regression from .2? | 10:42 |
jibel | cjwatson, other than this, I think we are good. | 10:42 |
cjwatson | (is this bug 813837? not much to go on there, it doesn't e.g. say *which* filename) | 10:43 |
ubot4 | Launchpad bug 813837 in ltsp (Ubuntu Lucid) (and 1 other project) "ltsp client not able to load boot file (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/813837 | 10:43 |
jibel | cjwatson, IDK, 10.04.2 tests are marked as passed. I'll try with a10.04.2 image | 10:43 |
jibel | yes, this one. I'm adding a comment and the workaround. | 10:44 |
cjwatson | I wonder if it's kernel-version-dependent - that's why knowing the filename it's failing to locate would be helpful | 10:44 |
cjwatson | apw: I'll look shortly when I have some I/O bandwidth (backups got a bit confused today and overran) | 10:45 |
jibel | not kernel dependent. | 10:45 |
jibel | the client tries to download the file ltsp/amd64.tmp/pxelinux.0 but the real location is ltsp/amd64/pxelinux.0 | 10:46 |
apw | cjwatson, as long as the team knows the status i am happy | 10:46 |
jibel | a quick workaround is to create a symlink on the server. | 10:47 |
jibel | anyone knows where this path is defined ? | 10:47 |
Daviey | cjwatson: Did skaet talk to you last night? | 10:48 |
cjwatson | yeah, I got a brain-dump when she was finishing up | 10:49 |
Daviey | The plan is/was to still release the current 10.04.3 candidate and release note the kernel issue. | 10:49 |
Daviey | (for server) | 10:49 |
Daviey | The fixed kernel will bake for 7 days in -proposed? | 10:50 |
cjwatson | https://wiki.ubuntu.com/LucidLynx/ReleaseNotes/ChangeSummary/10.04.3 says "if" for that - I thought it was certain? | 10:50 |
cjwatson | the alternative would be to do rapid-fire QA on it and waive the waiting period; but it's possible that that's too risky for the kernel, since it affects more than server if some unforeseen thing goes wrong | 10:51 |
Daviey | myself and robbie were OK with releasing 10.04.3 with the bug, and release noting it. | 10:52 |
cjwatson | ok, clearly your call :) | 10:52 |
Daviey | We felt the time (& expense) involved with re-QA'ing the image outweighs the benefit. | 10:53 |
Daviey | *ideally* i'd like the kernel in -updates on announcement, but that isn't the end of the world. | 10:53 |
Daviey | If this was a non-point release, it would be more serious IMO. | 10:54 |
Daviey | The bug already exists in the archive, so it's no differnet from 10.04 + updates. | 10:54 |
cjwatson | jibel: hunting ... | 10:56 |
jibel | its in /etc/ltsp/dhcp.conf | 11:00 |
cjwatson | dhcp.conf -> dhcpd.conf? | 11:01 |
cjwatson | I'd guessed that, but was trying to find what was doing the substitution | 11:01 |
cjwatson | (phone, though) | 11:01 |
jibel | dhcpd.conf sorry | 11:02 |
jibel | ltsp didn't change since the release of lucid, this bug must be there from the start. | 11:04 |
cjwatson | jibel: except I'm not sure that something else isn't substituting stuff into that file | 11:19 |
pitti | cjwatson: do you want lucid-updates stay locked down for now, or is it ok to release proposed packages? | 13:13 |
cjwatson | pitti: locked please | 13:16 |
cjwatson | I haven't taken a snapshot yet | 13:16 |
pitti | cjwatson: ack | 13:16 |
* pitti also tells the other SRU-eros | 13:16 | |
cjwatson | shouldn't be too long | 13:17 |
cjwatson | jibel: I didn't get much further tracking that down; the target architecture gets substituted in from debian/ltsp-client-builder.postinst, and this seems to indicate that /opt/ltsp/images/amd64.img.tmp didn't get removed | 13:19 |
cjwatson | but I don't know why | 13:19 |
cjwatson | jibel: at least wew have a workaround | 13:19 |
jibel | cjwatson, I'm stuck at the same place. I tried 10.04.2 which exhibits the same bug, I'll try 10.04. | 13:23 |
cjwatson | jibel: is there an oneiric system involved anywhere in this mix? | 13:23 |
jibel | and there's nothing interesting the the logs. mksquashfs exit correctly. | 13:24 |
jibel | cjwatson, no oneiric system involved | 13:24 |
cjwatson | odd | 13:25 |
cjwatson | well, if it's not a regression from .2, definitely not a blocker | 13:25 |
jibel | agree | 13:25 |
cjwatson | starting prepublication now | 13:26 |
cjwatson | would appreciate it if nobody else touched antimony :-) | 13:26 |
pitti | cjwatson: oh, btw | 13:27 |
pitti | cjwatson: there was a www.prev when I looked yesterday | 13:27 |
pitti | cjwatson: this might have been a leftover from me from alpha-2, unless you created it yesterday | 13:27 |
pitti | so please kill if it isn't your's | 13:27 |
cjwatson | already done - publish-image-set.py's commands do that | 13:30 |
skaet | good morning cjwatson, pitti - publishing started I see :) | 13:34 |
cjwatson | yep, though I may have to interrupt it to go out for a bit | 13:34 |
cjwatson | but I'll try to get prepublication done first | 13:34 |
skaet | cjwatson, fair enough. | 13:35 |
cjwatson | prepublication syncing out now - the 10.04.2 DVD is gone without being replaced by the 10.04.3 one because I got things in slightly the wrong order, but that'll be rectified in a few hours | 13:38 |
=== Ursinha` is now known as Ursinha | ||
skaet | cjwatson, just to confirm, we're not publishing powerpc images? (since no tests run on them.) | 13:39 |
=== Ursinha is now known as Guest44940 | ||
=== Guest44940 is now known as Ursinha_ | ||
cjwatson | skaet: correct | 13:42 |
skaet | coolio. | 13:42 |
skaet | :) | 13:42 |
cjwatson | point-release-snapshot running; I'm going out for a bit now, back in perhaps an hour, I basically have to wait for mirrors at this point anyway | 13:43 |
=== Ursinha_ is now known as Ursinha | ||
skaet | cjwatson, ok | 13:48 |
jibel | cjwatson, 10.04 doesn't have the bug, now trying 10.04.1 | 13:50 |
cjwatson | pitti: consider lucid-updates unfrozen now | 13:53 |
pitti | cjwatson: ah, splendid | 13:53 |
apw | cjwatson, did we publish the lucid kernel ? | 14:06 |
cjwatson | apw: no, not yet, I was going to do that once I'd finished with image publication | 14:29 |
cjwatson | if somebody else wants to deal with copying it, I'd appreciate it actually | 14:30 |
=== brendand_ is now known as brendand | ||
skaet | jibel, cjwatson, pitti, brendand - kernel candidate to fix 588861 has now been copied to -proposed. Tracking bug is bug: 813507 | 14:42 |
pitti | ah, thanks for the info | 14:43 |
brendand | skaet - we're not trying to release 10.04.3 with that version are we? | 14:46 |
* brendand thought 10.04.3 was released today | 14:47 | |
cjwatson | brendand: no | 14:47 |
cjwatson | 10.04.3 is locked | 14:47 |
skaet | brendand, we want to get it into -updates as soon as possible. | 14:47 |
=== jamespage1 is now known as jamespage | ||
jibel | cjwatson, I think I found the problem with ltsp | 14:55 |
jibel | in ltsp-client-builder.postinst the script exit too early from the while loop to update the debconf progress bar | 14:55 |
jibel | so dhcpd.conf is updated before the image is actually ready | 14:55 |
jibel | At this stage its name is $(arch).img.tmp | 14:55 |
jibel | then ltsp-update-image finishes normally and rename the image, but dhcpd.conf is already updated with the wrong name. | 14:56 |
jibel | This bug still exists in oneiric and is not always reproducible | 14:56 |
cjwatson | ah | 14:57 |
=== Ursinha is now known as Ursinha` | ||
=== Ursinha` is now known as ursinha` | ||
=== ursinha` is now known as Ursinha | ||
cjwatson | I can't really tell what's going on with mirroring since we forgot to start running the mirror prober more frequently, but it's had a couple of hours now so I think I'll just go ahead with the main publication | 15:52 |
apw | skaet, what was the decision on QA for the lucid kernel | 15:55 |
skaet | cjwatson, sounds good. | 15:56 |
skaet | apw, setting up a meeting to discuss what makes sense as a merge of the traditional SRU testing process and standard server testing. | 15:56 |
skaet | ScottK, just confirming we have your signoff to release the Kubuntu 10.04.3 images that have been tested on the iso tracker? | 16:19 |
ScottK | Let me look. | 16:19 |
ScottK | skaet: I'm good with Kubuntu desktop and alternate on i386 and amd64, yes. | 16:21 |
skaet | Thanks, ScottK | 16:21 |
cjwatson | skaet: all yours now, I've done everything up to point 3 on https://wiki.ubuntu.com/PointReleaseProcess (reload that if you have it loaded already, I've been editing it recently) | 16:41 |
cjwatson | I'll be around for a bit longer if you spot anything wrong | 16:42 |
skaet | cjwatson, thanks. | 16:44 |
cjwatson | I've re-enabled copy-report | 17:26 |
cjwatson | (so that should be automatically copying stuff from -security to -updates again) | 17:26 |
* micahg hugs cjwatson | 17:27 | |
brendand | skaet - what is the bug this -proposed kernel is meant to fix? | 18:06 |
cjwatson | brendand: bug 588861 | 18:43 |
ubot4 | Launchpad bug 588861 in linux (Ubuntu Maverick) (and 6 other projects) ""pad block corrupted" error when trying to register an image with 2.6.34 kernel (affects: 2) (dups: 1) (heat: 22)" [High,Fix released] https://launchpad.net/bugs/588861 | 18:43 |
cjwatson | it seems to be properly referenced in the changelog | 18:44 |
skaet | ScottK, was trying out the links, and http://ubuntu.mirrors.pair.com/releases//kubuntu/lucid/kubuntu-10.04.3-desktop-i386.iso is giving You don't have permission to access /releases//kubuntu/lucid/kubuntu-10.04.3-desktop-i386.iso on this server. Standard symptom of waiting for mirrors? or something else? | 18:45 |
ScottK | I'm guessing the mirror didn't sync yet. | 18:46 |
ScottK | http://ubuntu.mirrors.pair.com/releases//kubuntu/lucid/ still says .2 | 18:46 |
skaet | also looks like MD5 sums need updating on: http://www.kubuntu.org/getkubuntu/download | 18:47 |
skaet | ScottK, I'll give it another 1/2 hour or so then for the mirrors. Am seeing the ubuntu images on the mirrors now. | 18:48 |
ScottK | If you can give the the correct md5sums, I can update it. | 18:50 |
ScottK | skaet: Should be fixed. I left 10.04.2 for now since some mirrors will still offer those for awhile. | 19:26 |
skaet | ScottK, ok. I'm seeing them pretty clearly on: http://releases.ubuntu.com/kubuntu/10.04.3/, so imagine at this point its just a matter of time. | 19:27 |
ScottK | Yep. | 19:27 |
skaet | 10.04.3 is announced. :) | 19:41 |
skaet | cjwatson, steps 4,5 done, 6 not needed. if you're around, can you deactivate the 10.04.3 milestone? (I still appear to lack the ability :-/) | 19:45 |
cjwatson | skaet: done | 19:46 |
skaet | cjwatson, Thanks! | 19:46 |
ScottK | cjwatson: If you have a moment, would you please run the packageset update script again. I think we've got all the new packages in finally. | 19:52 |
skaet | heh, more automatic out-of-office/on vacation replies to the announce then usual. Yup, guess it must be summer vaction time. :P | 19:52 |
slangasek | cjwatson, skaet: congrats on 10.04.3! | 20:48 |
skaet | thanks slangasek. :) | 20:49 |
cjwatson | ScottK: done, http://paste.ubuntu.com/649464/ - anything else, drop me a mail with the exceptions that are currently in core/desktop-core and shouldn't be | 20:54 |
ScottK | cjwatson: Thanks. | 20:54 |
Daviey | skaet: is it too late to ask for a respin of 10.04.3? | 20:58 |
skaet | Daviey: yes | 20:59 |
skaet | it is too late | 20:59 |
Daviey | skaet: Oh good. | 20:59 |
skaet | :) | 21:02 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!