| ametheus | Guys, I'm suddenly not sure about bug 808498 | 00:51 |
|---|---|---|
| ubot4 | Launchpad bug 808498 in gnome-terminal (Ubuntu) "default LS_COLORS are annoyingly unreadable (affects: 1) (heat: 8)" [Undecided,Confirmed] https://launchpad.net/bugs/808498 | 00:51 |
| ametheus | It's either an opinion, or a tiny design fix | 00:52 |
| ametheus | what do you think? | 00:52 |
| ametheus | Well, serves me right for late-night bug management. | 01:01 |
| ametheus | Do drop me a line; I'd like some feedback on my bug gardening skills. | 01:02 |
| === pedro_ changed the topic of #ubuntu-bugs to: Ubuntu Bug Squad - next meeting 8/11/11 | http://wiki.ubuntu.com/BugSquad | Documentation: http://wiki.ubuntu.com/HelpingWithBugs | Want to report a bug? Read https://help.ubuntu.com/community/ReportingBugs | User support (not related to triage) is in #ubuntu | ||
| om26er | why do I see 'launchad bug tracker' confirming bugs these days? | 13:31 |
| pedro_ | om26er, http://blog.launchpad.net/bug-tracking/automatic-confirmation | 13:33 |
| om26er | pedro_, reading, thx :) | 13:35 |
| pedro_ | yw | 13:35 |
| === bambee_ is now known as bambee | ||
| === deryck is now known as deryck[lunch] | ||
| RedSingularity | jibel: bug 802409. Did you notice that on previous versions or just oneric? | 17:16 |
| ubot4 | Launchpad bug 802409 in update-manager (Ubuntu) "Do not propose to dist-upgrade without a network connection (affects: 1) (heat: 158)" [Undecided,New] https://launchpad.net/bugs/802409 | 17:16 |
| === deryck[lunch] is now known as deryck | ||
| bdmurray | Ampelbein: thanks for the idea regarding search-bugs and all of ubuntu | 18:14 |
| bdmurray | RedSingularity: what's going on in bug 798529? | 19:16 |
| ubot4 | Launchpad bug 798529 in update-manager (Ubuntu) "package update-manager-kde 1:0.150.2 failed to install/upgrade: ErrorMessage: subprocess installed post-installation script returned error exit status 139 (affects: 1) (heat: 8)" [Undecided,Incomplete] https://launchpad.net/bugs/798529 | 19:16 |
| bdmurray | RedSingularity: specifically what are you trying to find out? | 19:17 |
| RedSingularity | bdmurray: I was looking for the state of that package he has. I meant to ask for the output of 'dpkg -s update-manager-kde' Not sure why I used 'apt-cache' :/ | 20:03 |
| bdmurray | RedSingularity: right but why? | 20:03 |
| RedSingularity | bdmurray: TermLog was showing a config error with that package. | 20:04 |
| bdmurray | Setting up update-manager-kde (1:0.150.2) ... | 20:04 |
| bdmurray | Segmentation fault | 20:04 |
| bdmurray | dpkg: error processing update-manager-kde (--configure): subprocess installed post-installation script returned error exit status 139 | 20:04 |
| bdmurray | It looks to me like it crashed | 20:05 |
| * RedSingularity looking | 20:06 | |
| RedSingularity | bdmurray: hmmm, but it was in the process of installing, correct? | 20:07 |
| bdmurray | RedSingularity: yes update-manager-kde was in the process of installing and something crashed - what we really need is an apport-crash report not an apport-package report | 20:08 |
| bdmurray | Thank you for taking the time to report this bug and helping to make Ubuntu better. Reviewing your log files attached to this bug report it seems that a package failed to install due to a segmentation fault in application being used by the package installation process. Unfortunately, this bug report isn't very useful in its current state and a crash report would be much more useful. Could you try recreating this issue by enabling appo | 20:09 |
| yofel | bdmurray: you're obviously not using an IRC client that handles the IRC character limit properly :P (you got cut off "by enabling appo...") | 20:11 |
| bdmurray | I have no limits! | 20:11 |
| yofel | heh | 20:11 |
| RedSingularity | bdmurray: I believe I was going to have the user reconfigure the package with 'dpkg' and see if it cleaned it up. It seemed like an isolated case to me. I will ask for the crash log though. | 20:12 |
| bdmurray | There are a surprising number of segfaults during package installations | 20:12 |
| RedSingularity | Best way would be to enable apport? No other quick command to temporarily run a apport service? | 20:13 |
| bdmurray | 'sudo service apport start force_start=1' and then trying to install the same package again? | 20:13 |
| RedSingularity | bdmurray: and thats it! ;) | 20:14 |
| bdmurray | we don't people on stable releases enabling apport! | 20:14 |
| bdmurray | suprising = 143 so far | 20:15 |
| RedSingularity | bdmurray: that number just for natty? | 20:16 |
| bdmurray | no all releases | 20:17 |
| RedSingularity | bdmurray: having the user run that command will take him through the steps for a new report? If so you think its better to close the original or just dupe the 2? | 20:19 |
| yofel | new report, as update-manager is not what crashed | 20:20 |
| yofel | (IMO) | 20:20 |
| bdmurray | RedSingularity: no they'll need to repeat the operation too | 20:20 |
| bdmurray | RedSingularity: here is the complete comment | 20:21 |
| bdmurray | http://paste.ubuntu.com/650212/ | 20:21 |
| RedSingularity | bdmurray: ok thanks. Doing it now. | 20:22 |
| RedSingularity | yofel: What package assignment would be more appropriate in that case? | 20:26 |
| yofel | well, there are 3 things that can possibly crash looking at the postinst script: pycompile, which, dash. So one if them depending on the crash report (apport will assign it) | 20:26 |
| yofel | *one of them | 20:27 |
| RedSingularity | yofel: which log are you reading that from? | 20:32 |
| yofel | no log, the log said " update-manager-kde (--configure): subprocess installed post-installation script" -> post-installation script is /var/lib/dpkg/info/update-manager-kde.postinst (here for you to read: http://paste.kde.org/100309) | 20:33 |
| yofel | so 3 things can crash: pycompile from line 6, which from line 5, or the shell itself | 20:34 |
| RedSingularity | yofel: ohhhh ok. I never looked at that before. Thanks much! | 20:35 |
| yofel | RedSingularity: tip, for non-crash errors in the scripts (like exit status 1 which tells you nothing), it helps to add '-x' to the first line for debugging as that'll print what the script does. (ONLY for debugging, and mostly if you can reproduce it yourself) | 20:37 |
| yofel | since that requires editing of the script, nothing the general user should do | 20:37 |
| RedSingularity | yofel: so the -x would go under #!/bin/sh ? | 20:40 |
| RedSingularity | under that line i mean | 20:40 |
| yofel | either add 'set -x' or use '#/bin/sh -x' | 20:41 |
| yofel | er, '#!/bin/sh -x' | 20:41 |
| RedSingularity | And then run the program again or just execute the script manually? | 20:43 |
| yofel | manually would be best, or run dpkg --configure. dpkg -i would overwrite the script again. (so debugging failures on install can be tricky :/) | 20:44 |
| === yofel_ is now known as yofel | ||
| jmburgess | Hello! I'm new to debugging and just had a quick question. I've been using the Bug's need a bugwatch site to find some low hanging fruit to get started (http://reports.qa.ubuntu.com/reports/launchpad-database/unlinked-bugwatch.html) and as I go through them I want to check off the ones that don't have an appropriate upstream bug link in the comments...what is a good way to do that? | 23:18 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!