=== bdmurray changed the topic of #ubuntu-bugs to: Hug Day! https://wiki.ubuntu.com/UbuntuBugDay/20080318 | Ubuntu BugSquad | http://wiki.ubuntu.com/BugSquad | Documentation: http://wiki.ubuntu.com/HelpingWithBugs | If you have been triaging bugs for a while, please apply to https://launchpad.net/~ubuntu-bugcontrol/ - http://lists.ubuntu.com/mailman/listinfo/ubuntu-bugsquad [03:03] :) [04:37] Err - I've been gone for quite a while. Can I still close bugs as Fix Released without going through any procedures? [04:55] cge: Yes; as long as they *are* released :) === Arby_ is now known as Arby [07:06] First 3 bugs hugged [07:07] time for me to go to work now [09:41] oh yeah, happy hug day everyone! [09:46] :) === chuck_ is now known as zul [12:33] bdmurray: is http://paste.debian.net/51394 a sensible and well formed clue addition? [12:51] hello :D [12:52] Hey [12:52] Uhhh, Hug day! [12:53] Iulian, Yeah :) [12:54] * Iulian is going to squash some bugs. [12:58] hey Iulian! [12:58] that's the attitude ;-) [12:58] Hello pedro_ :-) === LifeHacker is now known as tuxmaniac [13:33] * pedro_ kicks cvs [13:39] pedro_: Git rocks ;) [13:39] you may want to try bazaar if you like git ;-) [13:39] but yeah it's good too [13:40] anything is better than cvs :-/ [13:40] i wonder why is still used on gstreamer [13:40] everytime i do an update from HEAD I've got ugly errors [13:40] that's everyday dammit [13:41] Yeah, that's true, I personally hate CVS. [14:02] Should I close this bug 189334? I don't think that's really a bug. What I can see from the logs is that his connection was interrupted. [14:02] Launchpad bug 189334 in update-manager "getting upgrade prerequisites failed" [Undecided,New] https://launchpad.net/bugs/189334 [14:02] (101 Network is unreachable) [14:03] Weird error from apt.log: Error: "/var/tmp/kdecache-pcnew1" is owned by uid 1000 instead of uid 0. [14:05] And that's the only error from apt.log. [14:07] Iulian: apart for the weirdness on the "owned by", it is clear the reporter had a network issue [14:08] Okay [14:08] so -- methinks -- go ahead and close it (and give him a nice explanation of why ;-) [14:10] Does anyone on bugcontrol think that this deserves a high severity? https://bugs.launchpad.net/ubuntu/+source/gnome-system-tools/+bug/198172 [14:10] Launchpad bug 198172 in gnome-system-tools "[users-admin] Changing anything about the root account sets and invalid home directory" [Undecided,Confirmed] [14:10] or indeed any other severity? [14:12] james_w: high is ok for it, I've already set it, thanks [14:12] james_w: my humble view: if this breaks system administration, it should be high [14:13] yeah, you can't use at least synaptic. [14:13] anything that needs $HOME for root really. [14:13] and the user doesn't even really have to do anything to trigger it. [14:14] mm is this another policykit + g-s-t breakage? [14:14] thanks pedro_ [14:14] seb128: have you seen bug 198172 ? [14:14] Launchpad bug 198172 in gnome-system-tools "[users-admin] Changing anything about the root account sets and invalid home directory" [High,Confirmed] https://launchpad.net/bugs/198172 [14:14] will ask garnacho in the meantime [14:14] pedro_: reading it now [14:14] I don't think that policykit get's involved at all. [14:14] better to ask garnacho [14:14] we don't have anybody really working on gst [14:14] ok ill do it and update the report [15:05] james_w: that looks great! I might use the url for 24626 and add it as a don't list. [15:08] bdmurray: great [15:08] bug 24626 is on apt, so I don't think it should be picked up here, is that right? [15:08] Launchpad bug 24626 in apt "Too many repositories "dynamic mmap ran out of room"" [Low,Confirmed] https://launchpad.net/bugs/24626 [15:08] something similar for apt would probably be good as well. [15:08] there's a few more for update-manager we could write [15:09] Cool, I'd like to clean up some of the ones we already have too. [15:09] In regards to apt its possible to have clue files inherit clues from other packages. [15:09] for instance it gets a lot when a package fails to install/upgrade, so something like "subprocess post-installation script exited with exit code" or whatever the message is would pick up a load that need reassigning [15:09] So maybe having update-manager inherit from apt would make sense for this clue. [15:09] bdmurray: ah, that's cool. [15:10] yeah, I think so. There's a few that would be duplicated across all these tools [15:18] Just tried using hugday-tools to mark off bug #173277 and get the error "main__.CloseError: 'Bug #173277 not found'" and hugday-tools doesn't seem to use malone so figured I'd mention it here [15:18] Launchpad bug 173277 in update-manager "Packages failed to install hardy" [Low,Confirmed] https://launchpad.net/bugs/173277 [15:19] asomething: thekorn, who just left, is the person you'll want to talk to [15:19] ok [15:21] bdmurray: btw, i applied to bug-control a few weeks back (via email and launchpad) with no reply https://edge.launchpad.net/~andrewsomething [15:24] asomething: I replied to your e-mail with the application did you get that mail? [15:24] hi secretlondon [15:24] hi qense [15:25] hmm... I guess I somehow missed it. [15:25] ready to hug some bugs? [15:25] hello [15:25] I'm opening the hug day page right now ;) [15:25] asomething: Shall I send it again? [15:25] When I believe I have fixed a bug and attached a debdiff is subscribung ubuntu-universe-sponsors the right thing to do? [15:26] should bug 107653 be marked as fix released or invalid? [15:26] Launchpad bug 107653 in usplash "ubuntu/ kubuntu 64 fails to boot on msi k9ngm" [Medium,Confirmed] https://launchpad.net/bugs/107653 [15:26] hi james_w [15:26] bdmurray: yes, please :-[ [15:28] asomething: sent [15:36] bdmurray: a.starr.b@gmail.com? [15:36] asomething: right [15:40] bdmurray: oh, it just showed up. It's the stock response which I replied to. Should I resend that reply? [15:41] asomething: nope, I found it. [15:41] Sorry about that [15:41] np [15:48] Does anybody know if there is a way to change the bug context in Malone w/o url hackery? [15:49] asomething: I'm reviewing it now. Did you notice gnome bug 517070 has a question in it? [15:49] Gnome bug 517070 in general "Last.fm Password Stored in Plain Text" [Normal,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=517070 [15:53] bdmurray: I did see that, but don't really know how to judge what the better way to do that is. A clean path would be better for the user as they wouldn't have to deal with any thing to get their password out of plain text but as the commenter said it would be kind of ugly. I was hoping that LastExit's dev might respond [15:54] I guess i should add that there [16:02] Isn't there a bug tag for gutsy2hardy? [16:02] i've been using it [16:03] there was a similar bug for FileZilla (passwords to servers stored in a plain text file) which was marked as won't fix by the developers citing "that is how we designed it" [16:04] qense: yes there is a tag like that [16:04] ok, thx [16:04] it isn't listed at the wiki page [16:04] shall I add it? [16:04] it should be just grabbing it from https://wiki.ubuntu.com/DebuggingUpdateManager [16:05] greg-g: is FileZilla part of gnome? [16:05] it isn't listened there [16:05] ah :P [16:05] it is [16:05] maybe they should be sorted [16:06] qense: a lot of the tags use wiki includes from a specific Debugging* page so we only need to update the tag in one location [16:06] ok [16:07] So I'm not sure it would be possible to sort by tag name. Maybe adding a navigation thing at the top would help. [16:07] So you could go to a specific table. [16:08] asomething: no, it is not. [16:12] Where do you request an upgrade of a package offered in the repositories? [16:14] file a bug against the package that needs to be upgraded [16:15] include a changelog and why the changes are necessary (specific bug fixes) [16:16] links to upstream announcments can help === hefe_bia_ is now known as hefe_bia [16:29] can bug 199603 be considered complete? [16:29] Launchpad bug 199603 in update-manager "Could not calculate the upgrade (7.04 to 7.10)" [Undecided,New] https://launchpad.net/bugs/199603 [16:29] the description seems pretty complete and the log files are there [16:29] Arby: yes, I happen to be looking at the same bug. I'm looking at the log files at the moment [16:30] OK, shall I set it as confirmed or leave it to you? [16:31] Sure! Since we have all the log files we need it is safe to confirm the bug report. [16:32] I'll set importance as low unless you think otherwise [16:32] I'm just looking for more details in the log file. [16:33] OK I'll leave that to you then [16:34] not sure about bug 199363 either [16:34] Launchpad bug 199363 in update-manager "Could not initialize the package information" [Undecided,New] https://launchpad.net/bugs/199363 [16:34] the description is pretty good but the reporter says there were no logs [16:35] mvo: does the cdrom upgrade ask you about using the network connection? [16:36] greg-g: thanks, I read past your comment [16:36] bdmurray: generally speaking it does, if the current version does I haven't tested yet [16:36] bdmurray: but it is designed to ask about a network [16:36] mvo: Would upgrading languages be a problem w/o network connectivity? [16:36] yes, definitely [16:37] pre-hardy it could very easily lead to a situation where the upgrade would not continue [16:37] Okay, that's what bug 199603 is about then. [16:37] Launchpad bug 199603 in update-manager "Could not calculate the upgrade (7.04 to 7.10)" [Undecided,Confirmed] https://launchpad.net/bugs/199603 [16:37] because it used to refuse to continue if trnalsations need to be removed [16:37] I relaxed this in hardy [16:37] thanks bdmurray [16:38] Arby: I'd be interested in seeing their /etc/apt/sources.list file for that bug [16:38] mvo: thank you! [16:38] ok I'll ask for it [16:39] thank you, you guys are doing great work [16:39] * mvo hugs the team [16:42] qense: no problem, let me know if you have any other questions [16:44] james_w: have you been using hugday-tools to update the wiki? [16:44] bdmurray: nope [16:45] I forgot about it :) [16:46] is huday-tools already in a PPA or does it still has to be compiled? [16:46] qense: it is python so doesn't need to be compiled, just pulled from bzr [16:46] ok, thx [16:50] got to go, bye [16:51] thekorn should set bugday-tools to use LP as its bug tracker. I was going to report a wishlist bug (simultaneously add the bug to your 5-a-day if you have that option set in the init) [16:52] s/bugday-tools/hugday-tools [16:55] * secretlondon is distracted dealing with gsoc`students.. [16:56] only done 1 bug so far.. [16:56] Boo :) [16:56] hi bddebian :) [16:56] Hi secretlondon [16:57] secretlondon: More than I've had time for :-) [17:03] Hi, I am running hardy(kubuntu) and experience a kind of system lock whenever i try to restart X or log out of it. No clue if that is a kdm or X or another problem. It happens on my machine(with fglrx) and on a friend's machine with intel graphics. [17:04] Can anyone point me to some bug description(if tracked, found none myself), or tell me wich logfiles are important in that case? [17:05] Exilant: I'd ask in #ubuntu-x as the developers there specifically deal with X [17:05] ok, i'll do so, thanks [17:10] should bug 201634 be closed? [17:10] Launchpad bug 201634 in update-manager "Can't install 'ubuntu-desktop'" [Undecided,Incomplete] https://launchpad.net/bugs/201634 [17:10] the user has since done a reinstall and the report is incomplete [17:14] got to go now, I'll check back later [17:19] thekorn: I want to submit a bug against hugday-tools but LP won't let me! :) [17:20] thekorn: I was going to report a wishlist bug (simultaneously add the bug to your 5-a-day if you have that option set in the init) [17:20] greg-g, works now, thanks for letting me know [17:20] oh, I'll report it, thanks [17:20] ok, cool [17:25] thekorn: since the latest 5-a-day release had you in the changelog, I assume you have links to the code for that. So I didn't supply them. [17:28] bdmurray: bug 182977 - just curious if you've tested recently [17:28] Launchpad bug 182977 in update-manager "distribution upgrade from dapper to hardy failed" [Undecided,New] https://launchpad.net/bugs/182977 [17:30] bdmurray: yeah, 182977 might just have been bad timing (when ubuntu-desktop was not installable for some reason - transition of packages etc). if you have the same image/machien a new test would be great [17:40] mvo: What's the right procedure for testing that again? [17:41] bdmurray: install update-manager from dapper-updates and run "update-manager -d" [17:41] eh [17:41] sudo update-manager -d [17:41] hopefully that gives you the little "upgrade now" button [17:43] mvo: It doesn't show up everytime [17:47] Arby: Yes, as there isn't sufficient information to determine what the bug was it should be closed as invalid. [17:48] bdmurray: thanks [17:48] no problem === secretlondo is now known as secretlondon [18:02] mvo: it's too late to make dialog changes for Hardy right? [18:04] yes, unless its somethng really important [18:05] bdmurray: hm, if it does not show up everytime, then I need to do a bit debugging, I will try to reproduce [18:07] mvo: they can be fixed in the "upstream" version of update-manager though right? I'm just wondering the right way to submit changes. [18:07] mvo: it would show up about 3 out of every 4 times [18:07] bdmurray: yeah, that would work [18:07] bdmurray: strange, that is a odd pattern [18:07] * mvo wonders if something is going on with a too narrow timeout or similar [18:15] ogasawara_: hurry up [18:16] I'm done biotch [18:18] Did you mean to mark 197015? I'm kind of working on it atm. [18:19] bdmurray: huh, so you are. I thought I had marked it confirmed bug a low priority [18:19] s/bug/but [18:19] heh - https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/197015/+activity [18:19] Launchpad bug 197015 in update-manager "[hardy] English strings about inconsistent package states wrong" [Undecided,In progress] [18:20] 10 seconds apart! [18:20] has anybody ever seen a "key lock"-bug in hardy? - just wrote a text and scrolled with my mouse wheel and all active text fields get filled with a's ,like I did not stop pressing the a-key [18:20] bdmurray: I win! [18:21] thekorn: there is a keys get stuck down bug [18:21] bdmurray: if you're working on that one you might want to do 196269 at the same time [18:21] secretlondon, ok, found it thanks [18:22] ogasawara_: okay, thanks [18:23] bdmurray: I looked into bug 200715, too and couldn't reproduce it either. [18:23] Launchpad bug 200715 in update-manager "Update Manager hang-up during recheck after update " [Undecided,New] https://launchpad.net/bugs/200715 [18:24] hefe_bia: I'd add that as a comment and ask them to test again [18:24] ok [18:27] I added some questions on #200715 - I wonder if it might be locale specific [18:28] mvo: With these string fixes that won't make it for Hardy should I do anything to the debian/changelog? [18:29] bdmurray: yes, I like it better this way, there will be merge conflicts then when I merge the fixes into my branch bevause the changelog has most likely changed since your fixes, but that is fine, I can reoslve them easily [18:33] mvo: would dch -e or dch -i be more appropriate? [18:33] dch -i is probably slightly better [18:34] Okay, thanks! [18:34] cheers [18:54] Looking at bug #199955. Reporter describes GUI lockups, but mentions these are also present in other programs. How would I respond? (I suspect the graphics driver or other broken packages) [18:54] Launchpad bug 199955 in update-manager "gtk? visual/input lockup at calculating new packages" [Undecided,New] https://launchpad.net/bugs/199955 [18:56] hefe_bia: I suspect its "just" the gui that is unresponsive, maybe ask him how long he waited? [18:56] hefe_bia: that is fixed in hardy, but it requires a updated python-apt unfortuantely, so for gutsy->hardy it still appears "hanging" [18:57] mvo: So he just has to wait longer for it to complete? Would it show some progress on the console if it was started in a terminal? [18:59] hefe_bia: no, the gui is unresponsive, depending on the system its just very few seoncds, but sometimes it can be longer. the console will not print anything too, its just the cpu that is busy that is the indicator that it works (and /var/log/dist-upgrade/apt.log growing too :) [19:00] ok, thanks. I'll as about how long he has waited and about the system (RAM, CPU speed) [19:01] thanks [19:01] he may also watch /var/log/dist-upgrade/apt.log [19:02] mvo: What status should I use for these bugs with a bzr branch? [19:02] hm, in progress or fix commited [19:03] okay and maybe milestone for Later? [19:15] is bughelper specific for the ubuntu project in Launchpad, or can I use it to look for bugs in arbitrary projects? [19:22] bug #181116 [19:22] Launchpad bug 181116 in terminator "terminal title options" [Medium,Fix committed] https://launchpad.net/bugs/181116 [19:26] TeTeT, you can use bugnumbers for all projects in launchpad, bughelper might be a problem [19:26] TeTeT: bugnumbers can definitely work with projects other than ubuntu, I don't recall about bughelper specifically [19:26] jinx! [19:27] hehe [19:28] thekorn: where do I get a valid cookie file from? the bugs are mostly private [19:30] TeTeT, you can use firefox 2/3 and epiphany cookies, just log in to launchpad with your browser once [19:31] thekorn: I thought I need to specify a cookie file with --cookie? [19:34] TeTeT, yes, for example if you used firefox 2 when you logged into your LP account give your ~/.mozilla/firefox/*/cookies.txt as the cookie argument [19:34] thekorn: on ff3 it seems to be cookies.sqllite [19:34] Regarding update-manager bug #200191: When there are no log files attached but from the description it is fairly obvious that it is a packaging problem, should I still ask for the logs or reassign to the package directly? [19:34] Launchpad bug 200191 in update-manager "System upgrade fails when dbgsym-package are installed (evolution)" [Undecided,New] https://launchpad.net/bugs/200191 [19:35] hefe_bia: feel free to do both, but always asks for the logs, the terminal log contians useful dpkg details [19:35] hefe_bia: I think that is probably an update-manager bug not a package bug [19:36] TeTeT, this sqlite cookie works with the latest version of bughelper in hardy [19:37] bdmurray: have you ever seen this again? [19:37] https://bugs.launchpad.net/ubuntu/+source/avahi/+bug/181822 [19:37] Launchpad bug 181822 in avahi "[hardy] avahi-daemon segfaults when booting LiveCD from 20080109" [Undecided,Confirmed] [19:37] Did it become pick on bdmurray day? [19:38] I'm grabbing the daily now to test. [19:38] bdmurray: yes, didn't you hear? [19:38] james_w: I have it around here [19:38] bdmurray: It's only a security update. Why should this need dist-upgrade? [19:38] * thekorn creates a "pick on bdmurray-day"-applet :) [19:41] james_w: honestly, I haven't checked recently [19:42] bdmurray: no trouble, just thought I'd ask [19:43] bug 196261 [19:43] Launchpad bug 196261 in update-manager "Could not calculate the upgrade" [Undecided,New] https://launchpad.net/bugs/196261 [19:43] should that be reassigned to open office [19:43] the apt.log shows that the probelm is a broken dependency in openoffice [19:44] thekorn: thanks, authentication works with the sqlite cookie. How do I write a clue file that matches all bugs? empty op? [19:44] Arby: which dependency? there was a well known bug regarding english language I think [19:44] TeTeT, do you just want to have a list of all bugs? [19:44] thekorn: yes, for start [19:45] TeTeT, then use bugnumbers, you won't need any clue-files [19:46] Package language-support-da has broken dep on openoffice.org-l10n-da seems to be the root of it [19:46] there are other errors referring to locale-da in various forms [19:47] Bugs like that can happen with the devel release when the archive is inconsistent i.e. maybe that one package had finished building yet but openoffice.org did. [19:47] thekorn: awesome, thanks [19:47] Arby: it has a distinctive openoffice flavour. However it doesn't really explain what the final problem that caused it to abort was. [19:48] what should I do with that report then? [19:48] it seems to be a bit of a moving target [19:48] Arby: it's probably worth just asking them if they can reproduce. [19:48] ok [19:48] yeah, I expect the problem is exactly the moving target. [19:50] TeTeT: you can do lots of interesting filtering with bugnumbers too [19:51] hi :) [19:51] hey afflux! [19:56] Können die Dateien aus /var/log/apt sensible Informationen enthalten? Bzw. kann man danach fragen, wenn jemand schon bei normalen Updates Probleme hatte? [19:56] Oh sorry, in english again: [19:57] Can files from /var/log/apt contain personal information? Can I ask for them in a bug report if people experience problems during a normal package upgrade (not dist-upgrade)? [19:59] hefe_bia: I think they contain no sensitive data, but I'm not sure if a simple upgrade is logged there [19:59] mvo: we talked about something similar before right? re permissions of /var/log/apt/term.log [20:00] meh, yes - my bad [20:01] bdmurray: wow, now that I wrote that small bughelper script for moving duplicates I suddendly needed it ~four times in two days :) [20:02] afflux: You're right. It seems to only contain manual package installations / upgrades. [20:03] Hm... does update-manager log normal upgrades? [20:03] hefe_bia: only in /var/log/apt/term.log [20:07] afflux: that's great! [20:07] a bug isn't occuring for the reporter anymore, but he says that he has changed his configuration... fix released or invalid? [20:08] Do we know how he changed the config? [20:08] bdmurray: I'm not really sure [20:09] his last comment in bug 178480 sounds like that [20:09] Launchpad bug 178480 in update-manager "Update errors in Gutsy Gibbon" [Undecided,Incomplete] https://launchpad.net/bugs/178480 [20:11] anybody speak French, trying to pick apart bug 195542 [20:11] Launchpad bug 195542 in update-manager "Impossible d'initialiser les données du paquet" [Undecided,New] https://launchpad.net/bugs/195542 [20:11] I can't really tell if it's a package problem or repos or something else [20:11] afflux: there might be enough to go on in the description to figure out what was wrong [20:11] I guess I just ask for the logs and see what they say [20:12] er, Arby [20:12] screen brightness does not change at all, even echoing to /sys/class/backlight/acpi_video0/brightness does not work. Is this a bug in the kernel? [20:14] er, did I just I answer my own question :) === blue-frog__ is now known as blue-frog [20:18] do the hugday tools work? [20:18] I get an error: __main__.CloseError: 'Bug #200191 not found' [20:18] Launchpad bug 200191 in update-manager "System upgrade fails when dbgsym-package are installed (evolution)" [Undecided,New] https://launchpad.net/bugs/200191 [20:22] next, should bug 194984 be set to wishlist, it's more a feature request than a bug [20:22] Launchpad bug 194984 in update-manager "Update Manager Doesn't Pay Attention to Network Status" [Undecided,New] https://launchpad.net/bugs/194984 [20:24] hefe_bia, argh, the issue now seems to be that the next hug-day on networkmanager has already been announced [20:24] thekorn: I specified --day 20080318 [20:25] Arby: I think that importance should be wishlist but a bit more could be done to make it more complete [20:25] hefe_bia, then it a bug, I will check tthis later [20:25] what else is needed? [20:26] I think the specific package version would be useful [20:26] just in case we don't get back to that bug in a while [20:27] thekorn: It's also not in the list when doing ./hugday list --day 20080318. Does it get the list from the webpage or directly from launchpad? [20:27] ok [20:27] hefe_bia, from the wiki-page [20:27] thekorn: strange [20:27] hefe_bia, it somehow misses the whole first block [20:35] mvo: bug 190782 adds transparent theming to update-manager and has a patch [20:35] Launchpad bug 190782 in update-manager "[Patch] RGBA support in update-manager" [Wishlist,Confirmed] https://launchpad.net/bugs/190782 [20:35] * mvo looks [20:36] bdmurray: hm, it seems to me like this should be done in the toolkit level (in gtk) [20:37] mvo: Okay, I've no idea. ;) I just happened to have it open in a tab. [20:37] yeah :) [20:37] thanks [20:37] thanks for looking === wolfger_ is now known as wolfger [20:41] where do I find thekorn's hugday script(s)? [20:42] secretlondon: https://bugs.launchpad.net/hugday-tools [20:42] secretlondon, https://code.edge.launchpad.net/hugday-tools/ [20:43] but it has a bug now, need to fix it [20:43] * secretlondon thanks the korn :) === heno_ is now known as heno [20:51] hefe_bia, I just pushed a fix, thanks for pointing this out [20:53] thekorn: hehe, I just had fired up my Python editor and also tracked it down to the regex ;) [20:54] python editor? you mean vi? *runs* [20:55] geany ftw! [20:55] ;) Actually I'm still using Komodo. It was my favourite IDE on Win.... I still like it though. [20:55] right, it's quite good [20:59] * greg-g is proud that he submitted the first bug report for hugday-tools :) [20:59] * secretlondon has been slacking today [21:00] secretlondon: me too, don't worry :) [21:01] of course, by slacking I mean writing papers and buying groceries etc instead of triaging ;) [21:01] I'm meeting and greeting tux4kid's google summer of code applicants [21:01] and eating pizza [21:02] not too bad of a day then [21:02] there are still some left for y'all ;) [21:04] I'll get to it on my next break from work [21:04] shouldn't /var/log/dist-upgrade contain a dpkg log file when upgrading from feisty to gutsy? [21:05] afflux: https://wiki.ubuntu.com/DebuggingUpdateManager mentions main.log, apt.log and term.log [21:06] k, thanks [21:11] I'll go to bed now, good night! [21:11] what importance should I set for bug 194469? [21:12] Launchpad bug 194469 in update-manager "Major crash on update from Ubuntu 6.06 to 8.04" [Undecided,New] https://launchpad.net/bugs/194469 [21:12] it looks quite nasty but I don't know if upgrading direct from dapper->hardy is even meant to work [21:13] Arby: LTS to LTS upgrades will be / is supported [21:13] dapper->hardy should work [21:14] ok should that bug be considered high importance then? [21:15] 194469 is a side effect of the python-central breakage we had a couple of days ago .( [21:16] it is a major failure, but it should be fixed with the recent python-central [21:17] mvo: so is confirmed,medium ok with you if a fix is available? [21:18] I added a comment, it should probaly be duplicated to the python-central bug that I forgot what number it was [21:28] thekorn: I get a bzr error after getting inventory 2/5 or something [21:28] Is it just me or does firefox in Hardy always say it needs to be restarted? I mean even if it is closed. [21:28] * secretlondon has no experience of bzr [21:29] mvo: is 194448 part of the same problem, apt-term.log refers to python-apt and pycentral again [21:29] bdmurray:I've never actually upgraded it without it being open [21:29] although not quite the same traceback [21:29] heh [21:29] Arby: was it at about the same date too? then that is pretty likely (sorry, haven't look at the output yet) [21:30] secretlondon, can you pastze the bzr error message somewhere [21:30] mvo: same day [21:30] thekorn: bzr: ERROR: Invalid http response for http://bazaar.launchpad.net/%7Eubuntu-bugcontrol/hugday-tools/main/.bzr/repository/packs/37bbe990e05b4a29b2cf422c2c1d0a4c.pack: Expected a boundary (h1XgKUvx:.Ym.:JdawV.) line, got '' [21:30] so probably the same problem [21:31] thekorn: it's repeatable, although I had to remove the directory to get it to try again [21:32] which version of bzr are you using? [21:32] i'll look, should be the current one though [21:32] trying to find the original pycentral bug [21:32] * thekorn hugs bzr for having the most cryptical error messages [21:33] thekorn the first time it said bzr: ERROR: Invalid http response for http://bazaar.launchpad.net/%7Eubuntu-bugcontrol/hugday-tools/main/.bzr/repository/packs/37bbe990e05b4a29b2cf422c2c1d0a4c.pack: Expected a boundary (0hTd(DmrjIZsi5H,h=?W) line, got '' [21:33] even cvs has never given me an error like that ;) [21:34] bzr 1.2~rc1-1build2 [21:34] secretlondon, looks like bug 198646 [21:34] Launchpad bug 198646 in squid "Invalid http response ... Expected a boundary" [Undecided,New] https://launchpad.net/bugs/198646 [21:36] secretlondon, but I have never seen this before, and I'm not a bzr expert [21:36] yeah could be caused by vodafone's hsdpa proxying/filtering stuff [21:37] I only ever use cvs and that's because its used in my upstream project [21:38] ok folks that's all I've got time for. enjoy the rest of hugday [21:39] did someone ask for a bzr expert? [21:39] * james_w runs off to find one [21:39] secretlondon: are you behind any sort of proxy? [21:39] it looks like the bug [21:40] james_w no idea but probably. I'm using vodafone hsdpa/3g internet [21:40] it has a fairly fascist kiddie filter etc [21:40] ah, I've heard your complaints about it before :) [21:40] :) [21:41] yeah, it's probably that bug report then (I didn't see someone had found it until now) [21:41] secretlondon: sorry that there is not much we can do. [21:41] sure [21:41] james_w, btw, do you know if there is any kind of documentation on how to integrate bzr into python scripts by directly using the bzr lib? [21:41] secretlondon: you could try the patch that John put in the bug report, but I know that's far from ideal. [21:42] thekorn: yes, there are a few things. [21:42] thekorn: there's a short document on writing a plugin in the source, but that doesn't apply to you too much I don't think. [21:42] thekorn: there is also http://bazaar-vcs.org/Integrating_with_Bazaar [21:43] secretlondon: there may be a way around this, but it would probably be *slow* [21:43] james_w, this link looks good, thanks [21:44] thekorn: if it doesn't cover something ask me (or #bzr if I'm not around) and we can help you and then update the page. [21:44] james_w: did you try using that update-manager clue file? [21:44] james_w: sure. I've added to the bug that I cna do a wireshark trace from my end if needed, obv I don't have access to v'phone internals [21:44] anyone have that python-central bug number handy? [21:44] bdmurray: no, I din't sorry. [21:45] secretlondon: ok, thanks, it's good to have a little noise on it. [21:45] james_w, ok, thanks [21:45] james_w: okay, I'll give a try then [21:45] I'm wary of ME TOOs as I get so many, but at least it shows its still a live issue [21:45] secretlondon: what might work is "bzr init --dirstate-tags whatever; cd whatever; bzr pull wherever" [21:45] bdmurray: thanks. [21:46] secretlondon: yeah, the usefulness generally decreases with the number, but adding something extra or offering some help can be good. [21:47] bdmurray: could apport integrate with bughelper at all, to point the user to similar issues, or is that pushing it a bit far? [21:47] james_w whatever and wherever are the same? =lp:hugday-tools [21:48] james_w: apport has some similar bug patterns that prevent bug filing [21:48] secretlondon: whatever is the path on your local disk, wherever is lp:hugday-tools [21:48] ah ok [21:48] bdmurray: ah, ok. [21:48] i'll try it [21:48] secretlondon: sorry for not being more specific. [21:49] james_w np :) [21:50] just checking, since I just started looking at the update-manager bugs. Does this look like the python-central bug: http://launchpadlibrarian.net/11946221/main.log [21:53] bug 192992 [21:53] Launchpad bug 192992 in python-central "[hardy] pycentral crashed with ValueError in parse_versions()" [High,Fix released] https://launchpad.net/bugs/192992 [21:54] thanks bdmurray [21:55] greg-g: if it is maybe we should write a clue for that too [21:55] james_w sorry for the newbie handholding but I now get bzr: ERROR: Not a branch: "/home/secret/bzr/lp:hugday-tools/". [21:58] secretlondon: ok, it looks like you'll have to spell it out, it appears to not resolve the lp: url for some reason. [21:58] i'll try that! [21:58] bzr pull http://bazaar.launchpad.net/~ubuntu-bugcontrol/hugday-tools/main/ [21:59] james_w back to the same boundary error [22:00] secretlondon: ah, sorry then, I think you are stuck. [22:00] * secretlondon nods [22:00] secretlondon: I can get you the code if you like [22:00] as a tar ball or something would be great [22:01] sure, one minute [22:03] greg-g: that is probably something different, the python-central bug usually happens during the upgrade (there is a term.log with content in the logs). this error here looks like a problem in the calculation of the upgrade, i.e. the upgrade has not actually be performred [22:07] mvo: my dapper to hardy upgrade seems stuck at trying to login after upgrading [22:08] bdmurray: oh? so that means, that the upgrade worked, you rebooted and now you can no longer login? [22:08] mvo: right, I saw that..which is why I haven't done anything with that bug yet :) [22:09] greg-g: ok :) [22:09] mvo: yes, it seems stuck at authenticating. time in gmd is incrementing. I think I can ping it. I can't switch to a tty - if I have the keystroke in vmware right. [22:11] secretlondon: http://jameswestby.net/scratch/hugday-tools.tar.bz2 [22:11] bdmurray: meh, that is bad. its not something silly like you assigned 192mb main memory and that is no longer enough under hardy (but was fine under dapper)? [22:11] bdmurray: did the upgrade itself perform ok, or was there a error? [22:12] mvo: you might be a winner I have 256MB of RAM. there was a notification about firefox.png. [22:13] bdmurray: ok, that comes straight from hell^Wthe gnome panel and should be harmless (also its something that should be easy to fix) [22:15] mvo: right and 256MB might not be enough correct? [22:16] bdmurray: it should, must it might not be, to be sure, just try with a bit more (384) and see if that helps [22:16] bdmurray: we used to be happy with 256, but I think stuff like deskbar-applet and the new printing applet eat quite a bit of memory [22:16] (both are python) [22:18] What do I do to a bug where the user says he doesn't have the problem anymore because of not using the mentioned packages anymore? [22:18] hefe_bia: Is there enough information in the bug report that someone could try recreating it? [22:19] bdmurray: There is an apport crashlog, but nothing more. [22:21] hefe_bia: which bug? [22:21] It's bug #189562 [22:21] Launchpad bug 189562 in update-manager "running partial upgrade could not calculate the update" [Undecided,Incomplete] https://launchpad.net/bugs/189562 [22:22] * secretlondon thanks james_w [22:23] hefe_bia: incidentally that isn't an apport crash bug but rather someone who went to Help -> report a problem in Firefox [22:24] secretlondon: no problem. [22:24] So there's nothing useful in there and it should be closed Invalid. [22:24] secretlondon: I can be your bzr proxy again if you need it, seeing as we broke it :) [22:25] james_ w thanks :) [22:25] bdmurray: ah, i see. I'm learning ;) [22:25] umm, how do you close a bug from a past hugday (actually today, but hugday-tool is trying to find it on the next one since the wiki page is up)? [22:25] greg-g: add --day 20080318 to the commandline [22:25] thanks [22:26] nope [22:26] hefe_bia: no problem. the tell for me was the tag "apport-bug" and "ProblemType: Bug" and "SourcePackage: firefox" [22:26] didn't work [22:27] apport-bug needs to be documented [22:27] that it's\not a crash but a "report a problem" [22:27] yup, the tag was what mislead me [22:28] thekorn: __main__.CloseError: 'Bug #191705 not found' [22:28] Launchpad bug 191705 in update-manager "failed dapper to hardy upgrade" [Medium,Incomplete] https://launchpad.net/bugs/191705 [22:28] secretlondon: its in https://wiki.ubuntu.com/Bugs/Tags where else should it go? [22:29] I dunno [22:29] Maybe in https://wwiki.ubuntu.com/Apport but that is mainly geared towards crash reports [22:30] I think adding an intro there before all the crash report stuff would be best though [22:31] Any takers? [22:33] bdmurray think the problem is that apport is seen as ==crash report [22:43] james_w: bug 181822 is fixed for me [22:43] Launchpad bug 181822 in avahi "[hardy] avahi-daemon segfaults when booting LiveCD from 20080109" [Undecided,Fix released] https://launchpad.net/bugs/181822 [22:43] is anyone still able to edit the hugday wiki page? [22:43] using the hugday-tool script [22:43] (sorry, need to complete my ideas) [22:44] bdmurray: thanks, I'll mark as incomplete. [22:46] ah, beat me too it. [22:46] I just wanted to let you know since you brought it up. ;) [22:47] mvo: still around? my upgrade to hardy seems to be in bad shape [22:49] bdmurray: yes, but about to leave. bad shape is bad! could you please file a bug and attahc the logs with a description of the problem (hangs on login is fine as a description) [22:49] bdmurray: I will look into it first thing in the morning [22:49] mvo: I've tried booting into single user mode w/o much luck either. I'll submit a bug then. [22:49] *ick* [22:49] single-user-mode does not work ?!? that is nasty [22:49] and no error during the upgrade? [22:50] Not that I saw, it's easy to try again. I think I'll install openssh first though. ;) [22:50] a request to limit the speed of updates (the bandwidth) is probably an issue again apt right? I mean, if it is implemented in apt then it will be in everything else (upgrade-manager, synaptic, adept) as long as they tie in the functionality [22:51] greg-g: that is true and already impleneted - but not exposed in the gui and really badly documented [22:51] s/again/against [22:51] greg-g: the documentaiton in only in /usr/share/docs/apt/examples/configure-index.gz [22:52] something like acquire::http::dl-limit or so [22:52] greg-g: so its a valid bug IMHO to make it easier available [22:52] bdmurray: Did you get bug 203385 when trying single user mode? [22:52] Launchpad bug 203385 in friendly-recovery "Recovery menu cannot be controlled with keyboard" [Undecided,New] https://launchpad.net/bugs/203385 [22:52] mvo: gotcha, thanks! [22:53] albert23: yeah! I sure did. [22:53] bdmurray: I have the same issue as you, also with lts upgrade [22:54] I was using kvm [22:55] albert23: It sounds like log files in /var/log/dist-upgrade/ would be helpful if it is an update-manager bug [22:56] bdmurray: I am still trying to see what's the problem. The system seems to be in quite bad shape. Sudo does not work, missing files in /dev .... [22:58] I see a lot of uevent: unsupoorted action string; messages [22:59] indeed, I see those as well [23:00] the system works a lot better when I boot with the old dapper kernel [23:17] bdmurray: if you give me a bugnumber before I go to bed, I promise to look at your upgrade bug first thing in the morning [23:21] Could someone have a look at bug #195818? My answer conflicts Sergio's because he answered while I was investigating... I'm pretty sure this bug is fixed in new python releases (Changelog mentions moving of site.py just after the problematic release) [23:21] Launchpad bug 195818 in python2.5 "Error whith dist-upgrade : erreur de traitement de /var/cache/apt/archives/python2.5_2.5.2-0ubuntu3_i386.deb" [Undecided,Fix released] https://launchpad.net/bugs/195818 [23:22] Since when can you add attachments when filing bugs? [23:23] its been on edge a while [23:23] mvo: bug 203756 [23:23] Launchpad bug 203756 in update-manager "using kernel 2.6.24-12 after upgrading from dapper to hardy isn't possible" [Undecided,New] https://launchpad.net/bugs/203756 [23:23] * hefe_bia reminds himself to hit reload before commenting ... [23:24] bdmurray: if you boot with 2.6.22-14 (or whatever old one is left after the upgrade) - does that help? [23:24] mvo: yes booting 2.6.15- works [23:25] this happened to someone else too [23:25] in /boot I see two initrd.img-2.6.24-12 [23:27] bdmurray: thanks, I added a comment, I'm sure its the udev package that was not upgraded, I will try to figure out wyh tomorrow. this is a pretty stock ubuntu install in dapper I suppose? [23:28] mvo: it should be clean from isvimage.ubuntu.com [23:28] bdmurray: cool, thanks [23:28] * mvo hugs bdmurray [23:30] albert23: did you see that? [23:31] bdmurray: yes I did. I have the same Held-back line mvo mentioned [23:32] Are you in a bind because of this at all? [23:32] bdmurray: sorry? [23:33] albert23: I remember you said it was a vm right? I was wondering if it was important for you to fix right now. [23:33] bdmurray: No, it's in kvm and I did it just for testing purposes [23:34] okay, great [23:44] bdmurray: I upgraded udev and now I can login in Gnome [23:51] is "network-manager" the backend or backend + gnome applet? [23:53] yuriy: there is network-manager-gnome and network-manager-kde [23:54] bdmurray: i meant the source package [23:54] i see it is both indeed [23:54] i mean, it isn't [23:54] * yuriy needs to read better [23:54] so hug day on thursday is for the backend then? [23:55] also, many knetworkmanager bugs should probably fall under network-manager [23:58] yuriy: I'd say yes to bother of your questions