[00:16] where can i get the packages for openoffice3 [00:16] intrepid [00:17] murdok: I believe there is a ppa for that [00:18] how can i search a packages in ppa? :$ [00:20] https://launchpad.net/~openoffice-pkgs/+archive [00:21] I used http://people.ubuntu.com/~kirkland/search.html by the way [00:22] :) [00:22] thanks [00:22] I have just found the same with: https://launchpad.net/ubuntu/+ppas [00:27] bug 276381 [00:27] Launchpad bug 276381 in offlineimap "offlineimap crashed with IOError in _display()" [Undecided,New] https://launchpad.net/bugs/276381 [00:37] greg-g: what's that? [00:38] just getting a link [00:39] I hate spamming the -bugs channel [00:39] * greg-g appologizes ;) [00:40] go hug a bug as punishment! ;) [00:40] sometimes, I wonder what I'm thinking when I recreate crazy bugs [00:41] what should be done with bug 154054? It's fixed in OOo 3.0rc2 but not in version 2.4.1. Should I open a bug upstream? [00:41] Launchpad bug 154054 in openoffice.org "[ooo-build] [hardy] OpenOffice 2.3 claims you can restrict permission for PDF using blank password" [Low,Triaged] https://launchpad.net/bugs/154054 [00:42] If its fixed in 3.0 I'd like for an already fixed bug upstream [00:43] Well, my question should be: will OOo 3.0 be in the repository of Hardy? [00:45] Hardy? definitely not, Intrepid? unlikely [00:46] But if they are preparing 3.0 version. [00:46] will they fix a bug in 2.4.1? [00:47] bdmurray, sorry I'm lost :-/ [00:49] Stable releases of Ubuntu only receive updates for security or bug fixes, not new versions of packages [00:49] So Hardy wouldn't see oo.o 3.0 [00:50] and it's too late for intrepid as well [00:51] as I understand the release date upstream has slipped and it was close to begin with anyway [00:51] okay, so to sum up: the only way to fix this issue in your machine is updating to OOo 3.0, right? [00:53] murdok: I'm not certain especially with comment 1 in that bug [00:54] yes that's strange because the label is present in 2.4.1-9ubuntu2 (current in intrepid) [01:26] good night all \o/ [01:26] murdok: with respect to OOo 3.0 it won't be releasing until at least Oct 14 so is too late for even intrepid [01:26] murdok: have a good night :) [01:27] just at time :p [01:27] okay [01:27] bye === chuck_ is now known as zul [01:56] I don't understand how to mark that you have hugged a bug on the wiki... I see no way to edit the page... am I missing something? [01:57] lifesaglitch: are you logged in? [01:58] greg-g: yes [01:58] greg-g: er [01:58] greg-g: no [01:58] :) [01:58] that is probably it then ;) [02:00] greg-g: You are a genius [02:00] is there a guide anywhere for adding a blog to the planet? [02:00] greg-g: Haha... I feel embarassed. I was signed into the launchpad and didn't realize it was a different system [02:01] lifesaglitch: happens more often than you think [02:01] lifesaglitch: but thanks for the compliment anyways ;) [02:01] mrooney: you an official Ubuntu Memeber? [02:01] Member, that is [02:02] greg-g: yes, indeed [02:02] mrooney: if so, this tells you https://wiki.ubuntu.com/PlanetUbuntu [02:02] greg-g: thanks! [02:02] mrooney: is this a new thing? (you being a member) [02:03] or am I just way out of the loop? :) [02:03] greg-g: when you sign in, does it not redirect to the page you were on? [02:03] lifesaglitch: it might not, unfortunately [02:04] lifesaglitch: it has been a while since I manually signed on, so I forget [02:04] greg-g: How do you do it? [02:04] lifesaglitch: I check the "remember me" box when signing in so it stores a cookie [02:05] gargh why are the vmware workstation downloads so big :( [02:05] mrooney: I see you were added to the Ubuntu Members team on Lauchpad back on Aug. 8th. Belated Congrats! [02:05] and always $#%#% up my network settings [02:06] jjesse: It's telling you to use VirtualBox or KVM. [02:06] wgrant: nice :) [02:07] greg-g: should I mark "trivial change"? [02:07] greg-g: thanks! :) [02:08] lifesaglitch: Only if you're changing something that doesn't affect the content. [02:08] ie. a formatting change would be trivial. [02:08] lifesaglitch: for just updating the Hug Day wiki, yeah, not a big deal [02:08] cuts down on someone's wiki mail [02:09] hmmm, can't "bzr checkout bzr+ssh://yourusername@bazaar.launchpad.net/~planet-ubuntu/config/main planet-ubuntu " be improved and greatly shortened now? [02:09] Hooray! I just hugged my first bug! [02:09] mrooney: bzr launchpad-login yourusername [02:09] bzr co lp:~planet-ubuntu/config/main [02:09] Wait. [02:10] That URL is very suspicious. [02:10] Errrm. [02:10] Whoever pushed that branch initially did a very bad job of it. [02:10] mrooney: it might [02:11] mrooney: I'm not sure to what though, something lp:~planet-ubuntu/config/main like [02:11] That's right. [02:11] Except the branch should be ~planet-ubuntu/planet-ubuntu/config or similar. [02:11] but to make sure it uses ssh and logs you in, not sure [02:11] greg-g: lp: does that if you've run bzr launchpad-login. [02:11] gotcha [02:14] I think I found the Ubuntu BugSquad poster: http://flickr.com/photos/kaet44/294528315/ [02:15] and it is CC:BY [02:15] in the stock responses, I have seen people post ...Unfortunately we can't fix it, , because your description... verbatim. Is a username supposed to go in the commas? [02:16] I would say no, just because that would add time to triagers' workflow and many would forget. If you want to start your repsonse with "User,/n" that's cool. [02:17] can you change that to only have on comma, please, lifesaglitch [02:17] I always start mine with "Hello %name, thanks for using Ubuntu and thanks for taking the time to file this bug report!" [02:17] greg-g: I was about to, but wanted to be sure before I did. :-D [02:17] lifesaglitch: cool, thanks! [02:18] mrooney: Does %name automatically populate their name? Or was that meant to be manually replaced? [02:19] lifesaglitch: no that is not automatic :) [02:19] mrooney: that would be a nice feature... [02:19] would be nice if %reporter_name did work though ;) [02:19] well there is a stock_replies thing for the lp greasemonkey scripts [02:19] so it could probably be added to insert a name [02:20] lp greasemonkey scripts? [02:20] file a feature request in launchpad! https://bugs.edge.launchpad.net/launchpad-project/+filebug [02:21] greg-g: I'm actually thinking both commas should be removed and one should be added after "Unfortunately" [02:21] I recently edited the stock replies to sound less harsh [02:21] it looks like someone reverted it poorly? [02:22] lifesaglitch: I agree. [02:22] I wonder why someone reverted my change :[ [02:22] is there a history? [02:22] mrooney: should be [02:23] mrooney: click the info link [02:23] oh it looks like bdmurray did it so, I guess it stands :] [02:24] but yes someone should fix the commas [02:25] Not sure I agree with his assessment [02:25] What can ya do though? [02:26] oh I see, it was only reverted for that one specific case [02:27] otherwise it seems to keep my "Unfortunately we can't fix it without more information." instead of "Unfortunately, we can't fix it because your description didn't include enough information." [02:28] I like the newer version better then [02:28] lol [02:28] The other sounds like it's saying "IT'S ALL YOUR FAULT YOU MORON" [02:28] :D [02:28] actually, to be a bit more precise, it should say something in the line of 'Unfortunately we cannot work on it without more information' [02:29] True [02:29] at this point it is unknown if it is something that needs a fix [02:29] Can we fix the user? [02:29] never [02:29] Sadness [02:29] we can, at most, work around them users ;-) [02:29] I just felt like the old (and current one in the top case) implies the bug is permanently unable to be fixed at first glance [02:30] Like, you get one chance, and you just blew it [02:30] It's not like we need users... can't we just make them all go away? [02:30] well... I hope not... they are the bane of support, and the reason [02:30] hggdh: +1 on the "we cannot work on it.." working [02:31] s/working/wording [02:31] yeah... I should have thought of that before... I think I myself edit this a long time ago, and did not notice [02:31] s/edit/edited/ [02:32] * hggdh is, meanwhile, having fun with coreutils [02:33] I have one claim to fame: I met the main author of "rm" [02:33] speaking of coreutils [02:33] wow! are you in California? [02:33] I was, for the summer. [02:34] you know who I'm talking about? [02:34] PHR [02:34] lowercase, more commonly [02:34] yes, but *I* never met them... a friend of mine did [02:35] hggdh: cool. He is a fun guy. [02:35] (while working at an outfit in CA, he got mad trying to write a kernel driver, and a bearded guy helped him...) [02:35] haha [02:35] phr actually doesn't have a beard [02:36] it was not phr, but one of the UNIX authors [02:36] or doesn't now, may have before. [02:36] gotcha [02:37] greg-g: lucky... lol [02:37] lifesaglitch: yeah, I am lucky sometimes. [02:39] does the commit actually go to the server? I thought you had to push after a commit, or is that only for branches? [02:46] mrooney: check to see if you can see your changes on Launchpad? [02:47] but that might only be for branches yeah [02:54] I assume someone would have noticed such a crucial aspect before me, so I will just assume it worked :) [03:08] good night all [03:09] it's actually good morning now, but i'm going to sleep :P [03:43] is there anyone here that does development work on imagemagick? The "convert" command seems to not be as good in hardy as it was in gutsy. Specifically, converting a ps (from a latex file) to a non-transparent png or jpg looks far more pixelated on hardy than in gutsy. [03:48] JohnPhys: No, the ImageMagick people are more likely to be in an ImageMagick channel than an Ubuntu bugs one. [03:50] ok [03:50] thanks [03:54] wgrant: you wouldn't happen to know what the imagemagick channel is? [03:54] JohnPhys: I have no clue, sorry. [03:54] #imagemagick is labelled as not have any developers. [03:55] So I presume it's not the right one. [03:55] indeed, just found that out [03:58] wgrant: Thanks anyway though. [04:04] alright, time to see if I successfully added myself to the planet [04:04] how long do new posts take to show up, anyone know? [04:04] a few minutes [04:04] probably around 10 [04:05] mrooney: you're on the list on the right, so that is a good sign [04:06] indeed! [06:48] good morning === philwyett_ is now known as philwyett [07:15] bdmurray: I noticed the boilerplate response on bug 251716. would it be possible to link to https://wiki.kubuntu.org/Kubuntu/Bugs/Reporting instead of https://help.ubuntu.com/community/ReportingBugs for kde bugs? [07:15] Launchpad bug 251716 in kdebase "Kicker crashes on any mount after startup" [Undecided,Incomplete] https://launchpad.net/bugs/251716 [08:24] good morning bugsquad [08:28] hiya thekorn [08:29] hallo dholbach [08:32] Hi I still have this anoying jockey-kde bug :( [08:34] bug #274357 [08:34] Launchpad bug 274357 in jockey "jockey-kde crashed with DBusException in call_blocking()" [Undecided,New] https://launchpad.net/bugs/274357 [08:34] can someone confirm this please? === dholbach_ is now known as dholbach [11:56] elmargol: see bug 274357 [11:56] Launchpad bug 274357 in jockey "jockey-kde crashed with DBusException in call_blocking()" [Medium,Incomplete] https://launchpad.net/bugs/274357 [12:15] Anyone running intrepid? I have a bug I marked as incomplete yesterday, but the user has confirmed the problem persists after Intrepid has been updated, can someone help me try and confirm this bug? [12:17] niadh: bug #? [12:17] 272094 [12:17] bug 272094 [12:17] Launchpad bug 272094 in openoffice.org "'Save-Discard-Cancel'-dialog isn't focused when closing with Alt-F4" [Undecided,Incomplete] https://launchpad.net/bugs/272094 [12:18] I tried intrepid in a vm, but couldn't recreate it [12:20] niadh: I can reproduce. I'll triage it :-) [12:21] cool, thanks, There may be a couple more I couldn't reproduce or see myself, but did as much as I knew to do [12:23] niadh: Yes, if you can't reproduce the issue by following the instructions, then the correct action is to ask for more information. [12:26] There's another bug with font rendering, but my eyes really can't see much different in font's, it's another intrepid issue [12:26] bug 271283 [12:26] Launchpad bug 271283 in openoffice.org "Open Office font rendering" [Undecided,Incomplete] https://launchpad.net/bugs/271283 [12:26] I would like assistance with https://bugs.launchpad.net/ubuntu/+source/openoffice.org/+bug/243271 [12:26] Launchpad bug 243271 in openoffice.org "Open Office Spreadsheet Sheet Copy error" [Undecided,Confirmed] [12:27] this must be a bug in openoffice, but I dont know how to progress now. [12:30] sysuser_: I marked it as confirmed yesterday, which means hopefully a developer will figure out the problem and fix it for us all :) [12:32] is it reported upstreams automagically, or does a ubuntu developer report upstreams? I found nothing on this bug in the OOo bug lists. [12:33] at openoffice.org [12:34] sysuser_: No, there needs to be intervention from someone to report it upstream. [12:35] sysuser_: However considering someone tried it with OOo3 and the bug is present I would imagine it would probably be an upstream bug. [12:36] is upstream reporting managed through launchpad, ur directly in the upstream bug databse, ie in the bug list at openoffice.org? (I reported OOo rc3) am I supposed to report it myself at oo.org? [12:43] sysuser_: Give me a moment to do a bit of research, I know we can link to upstream bugs, but I don't know if we do that to move the bugs upstream or simply because we find two of the same bug in two different bug report databases. [12:44] niadh: thank you [12:44] niadh: either really [12:45] james_w: So in this case, should we mark it as upstream and pass it up to the openoffice guys or not? [12:45] sysuser_: you can report your bug upstream, and then use "Also affects project" to link that to the LP bug report [12:55] Hew: Looking at upstream stuff left over from yesterday, this is in incomplete, but I believe it should be in upstream. [12:57] https://bugs.launchpad.net/ubuntu/+source/openoffice.org/+bug/40537 [12:57] Launchpad bug 40537 in openoffice "[Upstream] [hardy] OpenOffice cannot write to NFS files" [Unknown,Confirmed] [13:01] niadh: That bug is already upstream. I'm not sure why it was changed to incomplete since the bug is still open upstream. [13:02] Hew: Was gonna say, will turn it green and in fact what do I mark the bug as specifically. [13:02] ? [13:03] niadh: Looking at the activity log, it was Triaged in Ubuntu, but changed to Incomplete after the last comment. Upstream confirms the bug is still present in "beta 3" (which I assume is OOo 3) [13:04] niadh: I would put it back to Triaged, but you can set it to Confirmed. [13:04] ok [13:04] niadh: and since there may be some confusion in the report, would be useful to post a comment about the current situation [13:05] Hew: I would do if I know what the current situation was :P [13:05] niadh: Current situation: Upstream says the bug is still present in OOo 3 b3, Debian closed their bug due to inactivity (not fix released as shown by the remote tracker), so Ubuntu should be left as Triaged. [13:05] ah okies [13:05] niadh: I read the bug reports at OpenOffice and Debian, as well as the activity log [13:06] Ah, fair enough, will bear that in mind for future reference. [13:06] niadh: Do you understand the layout of launchpad and how the upstream links work? [13:07] Hew: I might do if I played about with upstream links, I just saw a bug in incomplete that had [upstream in the title] so I looked into if it was actually in the right place more than any real knowledge of what to do with it. [13:09] james_w: i will try to do that. thank you james/niadh. [13:09] sysuser_: Your welcome. :) [13:10] niadh: [upstream] and [$ubuntu_release] in the title appears to be something the openoffice team do. Looking at the top of the bug report itself is how you really find out what's going on :-) [13:10] Hew: Wow that bug is ancient. [13:11] It's been kicking around since 2005 [13:11] niadh: as ancient as bug 1? :P [13:11] https://bugs.launchpad.net/ubuntu/+bug/1 (Won't display info) [13:12] Yeah but bug 1 is marked as fix in progress ;) [13:12] https://bugs.launchpad.net/ubuntu/+bug/1 (Won't display info) [13:12] hehe === pedro__ is now known as pedro_ [13:28] hum. that was silly. === Ziroday is now known as ziroday === LucidFox_ is now known as LucidFox === thunderstruck is now known as gnomefreak === thunderstruck is now known as gnomefreak === thunderstruck is now known as gnomefreak === bdmurray changed the topic of #ubuntu-bugs to: Next Hug Day! is Thursday, October 2nd | 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/ | Want to report a bug? Read https://help.ubuntu.com/community/ReportingBugs | User support (not related to triage) is in #ubuntu === thunders1ruck is now known as gnomefreak === thunderstruck is now known as gnomefreak === asac_ is now known as asac [16:56] what a bizarre bug, bug 276744 [16:56] Launchpad bug 276744 in openoffice.org "OO.org Calc hangs when Ctrl-C, Enter, Ctrl-Z, Enter key sequence pressed" [Low,Confirmed] https://launchpad.net/bugs/276744 [17:20] Boo [17:42] the ubuntulove tag, that's a new one to me [17:43] mrooney where? [17:45] leoquant: bug 163206 [17:45] Launchpad bug 163206 in xubuntu-default-settings "[Xubuntu only] Non-translatable labels and launchers" [Low,Triaged] https://launchpad.net/bugs/163206 [18:24] Anyone wanna start a discussion regarding (IMHO anyway) the lack of backporting bugfixes into LTS versions? [18:24] I'm not quite sure of the policy and wanted to get more information [18:25] Well backports are usually actual new versions of software, right? [18:25] mrooney: is that what it's considered? [18:25] maybe my wordage is wrong [18:25] Right so pidgin 2.5 was a Hardy backport [18:26] ok - so what if a newer version fixes critical issues in software though? [18:26] but we can also cherry pick patches for any prominent bugs in 2.4 and put them in -updates [18:26] ok [18:26] It probably doesn't happen as often as it could [18:26] I'd like to get more involved with this process [18:26] IMHO it doesn't, yes [18:27] I run a fairly unique network(s) with LTSP under Hardy [18:27] but from a developer standpoint, let's say you fix a bug in 1.1 of an application, repackage, and get it into say Intrepid [18:27] and there are bugs with software that affect only LTSP terminals [18:27] ok [18:27] now, anyone can upgrade and get that fix. Does it make sense to put your limited time into making it available for previous versions, or fixing more issues for everyone? [18:27] I think part the issue is limited people and time [18:28] IMO it makes more sense to work on issues that aren't fixed at all given limited time, instead of patching older versions of software which are already fixed in newer versions [18:28] mrooney: Maybe that's the issue then, yeah - I think it would make MORE sense to first fix it for the LTS version than Intrepid [18:29] But that order is rarely possible [18:29] how come? [18:30] With pidgin for example. those developers fix and issue in 2.5, Intrepid (or the next release when the bug is fixed upstream) now in sense already "have" that bugfix since that is the version they will get [18:30] as soon as the bug is fixed future versions essentially are "fixed" instantly, you can't fix it first somewhere else, do you know what I mean? [18:31] not really :) [18:31] i mean i understand the new versions containing bugfixes upstream [18:31] so once you fix and it know, okay all future versions will have this issue addressed, should I a) fix more bugs or b) apply this fix to older versions [18:31] Oh, ok [18:32] well...I think then, that LTS shouldn't be touted as a "stable" version then [18:32] because in all respects it really isn't, if you're not fixing bugs as a priority along with security updates. [18:32] well, it is the non-changing aspect of it that MAKES it stable [18:33] Lns - I just wanted to point out that LTS != stable [18:33] that doesn't make sense to me at all...just because a buggy program doesn't change doesn't make it stable ;) [18:33] a lot of people mis-understand that [18:33] all releases are stable [18:33] chrisccoulson: along with me =) [18:33] Lns: https://help.ubuntu.com/community/UbuntuBackports might help [18:33] LTS means it is supported for longer [18:33] ok...lemme read that link real fast [18:34] Lns: on the other hand though if an issue fixed in Intrepid is affecting a lot of Hardy users, the patch can be and sometimes is applied [18:34] Lns - this link explains the criteria for updating packages in stable releases: https://wiki.ubuntu.com/StableReleaseUpdates [18:34] yes that is probably a better link [18:35] chrisccoulson: ty =) [18:35] you're welcome [18:36] Lns: so basically if you know of a bug that was fixed in a newer version and want it in the current Ubuntu release, you can request the patch to be applied and an SRU done, by filing a launchpad bug [18:36] chrisccoulson: would you say that accurately describes the process? [18:36] So...ok. I'm understanding a bit more here now...but what I still don't get is, if I want a stable version of Ubuntu (as in bugs are fixed) which one do I choose? And don't say Intrepid because that's beta and I've gone down that road before =) My situation is that I run many LTSP-based servers with at least 35 terminals on it at the same time, and things like Firefox and OOo have bugs that seem to only be fixed in the latest/greatest versions, which have HU [18:36] GE issues in other parts of the system that make it unusable in a production environment [18:37] mrooney: that sounds like a process i'd love to get involved with [18:37] Lns: right, I think we need people like you to identify such bugs [18:38] mrooney: I have an army of people to help me =) [18:38] mrooney: what's the def. of an SRU? [18:38] stable release update? [18:38] yeah, that is the wiki page chrisccoulson linked you to [18:38] oh..haven't looked at it yet ;p [18:38] heh [18:38] oh yeah I think his link will clarify things more than mine [18:39] Awesome. [18:40] I need to run for the moment but i'll probably be back with more questions later...thank you guys for the help, I'm excited about helping out with this [18:40] but so in the case of pidgin again, 2.5 came out and tons of bugs were fixed, right? but many of those might not even be in the ubuntu tracker, and even if they were it would be hard to link them all to upstream trackers and know that they were fixed in a new version [18:40] and that is just one of thousands of packages [18:40] so without people telling us "hey, this issue was fixed, can you apply the fix to Hardy?", we don't know, but it can definitely be done if we do [18:40] okay, bye for now! [18:41] mrooney: so it's a matter of just communicating that, huh? [18:41] Lns: that is a large part yes, and of course meeting the SRU requirements [18:41] right [18:41] some trivial issue that has a chance of breaking things for lots of people, or a default behavior change, probably isn't going to make it [18:41] well i see a lot of potential for different use cases of Ubuntu for keeping on top of that kind of thing for LTS releases [18:42] yeah I agree [18:42] can't these newer packages be provided using other repos? [18:42] I'll bbl..thanks guys [18:42] Hamra: yeah, if users want to set up PPAs for packages of concern or just download new debs, that can happen [18:43] backports? ppa? or something? [18:44] Yeah but Lns was talking about getting more bugfixes in through -updates for stable releases [18:45] ah, i see [18:45] backports is really for new versions [18:46] it's a lot better if we have more such updates. of course, after thorough checking. but then again, 6 months for new releases isn't much [18:47] Yeah that is what I mean, if you see a bug fixed and know, okay the new Ubuntu will have that in a few months, do I really want to spend time making a patch and going through the SRU process to get it into the current release [18:47] but now that i know how triaging work, we have to admit that we are having a *huge* problem with lack of triagers [18:48] there's just a *big* number of bugs raining on us daily [18:51] yes, that's true, I don't think it is as bad as a lot of people initially think, but at the same time definitely more people helping in almost every area could do a lot of benefit [18:51] on another note, does anyone know how to debug freezes when returning from TTYs? [18:52] it goes back to X and shows a mouse cursor but always freezes instantly and I have to use the magic keys to restart [18:53] is the keyboard or mouse by any chance USB? [18:55] I don't think, this is on a laptop [18:55] my internal cd-rom drive is USB oddly enough [18:56] laptop? never mind usb then [18:58] isn't there a way to set xorg to log to a file that you can recoverlater? [19:05] https://wiki.ubuntu.com/X/Debugging [19:13] bdmurray: thanks! now I just need another machine [19:13] Lns: http://blogs.gnome.org/seb128/2008/01/28/ubuntu-stable-updates/ may interest you, as well [19:14] but isn't the X log file overwritten upon reboot? [19:29] Hamra: well you log in remotely from another machine and grab it or the backtrace [19:29] Lns: also, http://laserjock.wordpress.com/2008/08/07/sru-needs-you/ [20:07] if someone is asking for a new version of some program, what should i do? tell him it's a wishlist? forward him to some other place? mark it invalid and explain why? [20:27] Hamra: wishlist, will happen at the start of Jaunty development probably [20:28] you're talking about the gimp bug, right? [20:29] Hamra: No, I'm responding to your last question regarding what to do about new version requests. [20:29] ah ok === mvo_ is now known as mvo === txwikinger1 is now known as txwikinger === apachelogger_ is now known as apachelogger === Adri2000_ is now known as Adri2000 [21:08] is it always so quiet in here? [21:38] it can be [21:42] can anyone confirm or deny in Intrepid that switching to TTY1 and then back launches the Gnome Help and also leaves the window in Move mode? [21:43] basically, it is like all the keys involved in switching also get sent to X when you return [21:43] ls [21:43] Yes, look at that. [21:44] is that...expected? [21:46] All my keystrokes get sent to X === ivoks2 is now known as ivoks [21:50] mrooney: only some of mine keypresses in a console end up happening in X [21:50] mine did not, at least none that I could see [21:51] bdmurray: is that expected? [21:51] mrooney: I'd never tried before Intrepid but I wouldn't think so [21:51] ls [21:51] I notice it also includes everything but the first two characters of my password, as well :| [21:51] hehe I think hggdh duplicated it [21:52] jeezz... I just got a xchat help pop-up after pressing F1 on TTY1... [21:52] Yup :) [21:52] cool [21:53] I can't seem to find a bug on it, what package might it be? [21:53] look at it as an improvement -- you do not need to be under X to work under X [21:53] and everyone in IRC can watch your commands! [21:53] what transparency [21:53] :-D I am an open book [21:53] this is actually quite bad [21:55] every key press? [21:55] I will go back to TTY1, and run a 'ls -l' [21:55] my X focus will remain on xchat [21:56] I think it is only some special keys [21:56] the same in KDE!!! [21:56] ls -l [21:56] lol [21:56] there you go... my 'ls -l' got sent to xchat [21:56] that's realy serious [21:56] and dangerous [21:56] it took a while, though, and seemed to happen only after I returned to TTY1. I will try again [21:57] Hamra: and thirdly hilarious! [21:57] pwd [21:57] hggdh: yeah, the input gets sent after you get back, for me it seems to take a click [21:57] yes [21:57] so there is a real bad mix here [21:57] now... who is responsible for that? [21:58] mrooney, are you running uvesafb? [21:58] how might I ascertain that? [21:58] lsmod | grep uvesafb [21:58] could it be because of the new input method? input got moved from X to... what's it's name? [21:59] hggdh: well, the result was two lines, does that mean yes? [21:59] console-setup? [21:59] this was *not* under TTY1 [21:59] mrooney, yes [22:00] there it is, input-hotplug! [22:00] could it be the reason? [22:01] Hamra, are you taliking about evdev? [22:01] yes [22:01] mrooney, did you open a bug on this? [22:01] no I was waiting to see what package [22:02] so that I might better search for a dupe [22:02] xkeyboard-config? [22:03] it seems too egregious to not have been filed but you never know, it could be new I suppose [22:03] Some detailed steps to recreate it would help, so others can experiment too. [22:03] okay, I'll file it under xkeyboard-config [22:04] bdmurray, how do you classify this? I think this is a serious issue [22:04] hggdh: I haven't been able to personally recreate so would like some more details [22:05] kees, lets wait for mrooney to file it [22:05] dammit [22:05] imagine someone is chatting, and for some reason, is using TTY1 to do some stuff, it could make a real nice conversation [22:05] I *always* get caught in the auto fill :-( [22:05] Yeah I was going to mark it security, if bdmurray agreed [22:06] mrooney: seems reasonable to me [22:06] Hamra, you reproduced it under KDE, right? [22:07] yes [22:07] so... gnome and KDE affected... should be common code... X [22:07] mrooney: i'm back fyi. Looked at those links, and thank you. I posted to the LTSP-Discuss list so others might be able to help out as well with getting bugfixes into hardy. [22:08] Lns: did you see the one about SRU needs your help? it seemed like it might be relevant to you? [22:08] mrooney: yes, that's the one i posted. [22:08] Directly relevant. That's exactly what i was looking for - it's hard to get the info on these processes, it's such a huge thing - but i'm getting it better now =) [22:08] ore [22:09] OK. I typed 'pwd once more', and only 'ore' got echoed in xchat [22:10] and it shows up right away when you switch back or ...? [22:10] i can't see to be able to send text from TTY1, but the function keys are sent to X [22:10] seem* [22:11] Lns: yeah, it can be hard to find out exactly what you want, there is so much going on, luckily we have a good community to help out so it makes all the bridging possible [22:12] for me it seems to work this way: 1) crtl/alt-F1; 2) type 'whatever', hit enter; 3) crtl/alt/F7; 4) crtl/alt/F1; 5) crtl/alt/F7, and I see the echo in xchat [22:13] does not seem to work on TTY2 [22:13] TTY1 [22:13] nano [22:13] bug 276887 [22:13] Launchpad bug 276887 in xkeyboard-config "Input to TTYs goes to X after returning" [Undecided,New] https://launchpad.net/bugs/276887 [22:13] it sent text! [22:14] I wanted to add another case about text being copied but all my TTYs are now broken :) [22:14] TTY1 [22:14] so I can write it exactly, maybe hggdh or Hamra can do it [22:15] correction: when I get back to X, as soon as I press CRTL/ALT the text is echoed on xchat [22:15] mrooney: yup, for sure. It seems fairly cluttered, but just by what changes I see in LP, things are getting much more consolidated. [22:17] hggdh: yeah, I think it has to do with the window being put in the "move" mode with you return; as soon as you exit that the input is sent [22:17] is anyone else editing it now? I am adding a second test case [22:18] go ahead [22:19] yes, I noticed at one point that the window was moving, but I could not yet repeat it [22:19] i'm adding a comment as well [22:20] TTY2 [22:20] TTY2 is also involved [22:22] mrooney: can you try it w/o compiz enabled? [22:22] bdmurray: sure [22:23] hggdh: so odd [22:23] should metacity --replace be sufficient? [22:23] is everything typed repeated in X? [22:24] kess, we do not know yet if everything. [22:24] kees: it doesn't seem like everything, just everything up to the second to last newline [22:24] BTW, I do not run compiz [22:24] mrooney: does this happen in Hardy? [22:24] (I'm trying to imagine what's changed) [22:25] n [22:25] vwxyz [22:26] for me it seems to be the last few characters, varying from one to (so far) four [22:26] kees: I am on Intrepid, I can't tell you for Hardy, since I can't return from TTYs on Hardy! [22:26] five, I mean :-( [22:26] kees: the only reason I found this is because I wanted to see if the that bug where I couldn't return from TTYs was fixed in Intrepid [22:27] it was but, then I found this :) [22:27] whoa. okay, reproduced here, but no idea about the cause yet. [22:27] bdmurray: metacity seems to perform exactly the same way, I will note in the description [22:28] kees, we opened it against xkeyboard-config, but we really do not know [22:28] I got 3 codes, 2 leaked from the tty, one unknown. I had typed into the terminal several nonsense lines, then "line 4", returned to X, saw nothing. then the instant I moved my mouse, I got "4" [22:29] h [22:30] h [22:30] bryce: we're examining https://bugs.edge.launchpad.net/ubuntu/+source/xkeyboard-config/+bug/276887 [22:30] Launchpad bug 276887 in xkeyboard-config "Input to TTYs goes to X after returning" [Undecided,Confirmed] [22:30] okay, I updated it to reflect compiz/metacity and Confirmed it based on input [22:30] I just typed 'whoami' on tty1; got back to X, right ctrl/alt, got 'h' echoed in xchat [22:30] bryce: stuff typed on the tty is leaking into X [22:31] hmm [22:31] I saw something akin to that with my sleep key - it doesn't work when I'm in X, but if I switch to ttw, hit it, and then go back to X, it will sleep [22:31] bryce: but not all of it, and not exactly what was typed. for example, I switched to tt1, and typed out the alphabet without hitting enter, switched back and got "uvwxyz" followed by alt-F7. ?! [22:31] bdmurray saw something a bit similar as well with his sleep key [22:32] kess, I confirm you last test [22:32] sleep == suspend, but isn't that an ACPI event? [22:32] kees, we opened it against xkeyboard-config, but we really do not know [22:32] kees: afaik [22:33] now, THAT was weird... I got a whole old sentence repeated here [22:33] hmm, I'm not reproducing it trivially, let me try another machine [22:34] nope [22:35] bdmurray also could not reproduce... [22:35] hggdh: I've been able to see the bug now [22:35] bryce: I put my reproduction notes in the bug report. [22:36] I see it in all ttys, even ones not running getty (ctrl-alt-f8) [22:38] I wonder if this is somehow console-kit? [22:38] ooh, I reproduced it [22:39] kees: that's what i thought, since intrepid differs from hardy that it uses input-hotplug, and not those values in xorg.cong anymore [22:41] yeah sometimes it prints part of what I typed, sometimes it prints just ~, sometimes random stuff like ;5~ [22:41] heh, that looks like a funky smiley [22:41] bryce: there does appear to be some level of speed involved -- not sure how to predict how much it'll spit out [22:42] bryce: same here, i just typed characters i'm not even sure i know in what language they're used! :P [22:43] it just* [22:43] kees: am I guessing correctly that this is a kernel issue, that the console mode ought clear whatever buffer this data is in, before going to X? [22:43] s/going to/going back to/ [22:43] bryce: for me I have a delay between returning to X and making a mouse or keyboard event [22:44] i.e. I return to X, nothing has happened, then if I move the mouse or touch a key, the leak appears followed by my recent mouse/keyboard action [22:44] let's see what xev thinks of all this [22:46] has anyone else noticed that it seems to kill the caret, if that is the right word, of gnome input widgets? [22:46] for example if you had a gedit tab or pidgin conversation focused, after focusing you can't get a caret back in the text field [22:46] it also depends on the speed of returning to X and moving the mouse [22:46] if i go straight away, i get a lot of characters leaked [22:47] wait 10 seonds, and i get only a tilde ~ [22:47] keycode 67 = (keysym 0x1008fe01, XF86_Switch_VT_1) [22:47] keycode 67 = (keysym 0x1008fe01, XF86_Switch_VT_1) [22:47] keycode 67 = (keysym 0x1008fe01, XF86_Switch_VT_1) [22:47] keycode 67 = (keysym 0x1008fe01, XF86_Switch_VT_1) [22:47] keycode 67 = (keysym 0x1008fe01, XF86_Switch_VT_1) [22:47] keycode 37 = (keysym 0xffe3, Control_L) [22:47] keycode 64 = (keysym 0xffe9, Alt_L) [22:47] keycode 67 = (keysym 0xffbe, F1) [22:47] keycode 44 = (keysym 0x6a, j) [22:47] keycode 44 = (keysym 0x6a, j) [22:47] keycode 44 = (keysym 0x6a, j) [22:47] keycode 44 = (keysym 0x6a, j) [22:47] keycode 44 = (keysym 0x6a, j) [22:47] keycode 44 = (keysym 0x6a, j) [22:47] keycode 64 = (keysym 0xffe9, Alt_L) [22:47] keycode 64 = (keysym 0xffe9, Alt_L) [22:47] it pint a lot of keycode 67 in there; dunno why [22:47] someone should be using pastebin [22:47] :) [22:48] I switched to tty 1, hit a bunch of j's and that's what came up [22:48] hrmph [22:48] bryce: yeah, I saw the same. xev agrees: it doesn't appear until i move them mouse, and then all the events fire [22:48] maybe the buffer is being cleared but slowly ? :S [22:49] kees, I can repeat by pressing CTRL/ALT [22:49] hggdh: what do you mean repeat? [22:50] I can get it to leak [22:50] bryce: do you still want the output requested in the comment, or are you okay since you reproduced it? [22:51] mrooney: yes, please attach it. It's almost always needed with X bugs [22:52] bryce, do you want more than one> [22:52] ? [22:52] hggdh: couldn't hurt. [22:52] roj [22:56] bryce, done [22:57] anyone know roughly how long it's been doing this? [22:58] i don't think anyone here ever noticed this beofre today [22:58] bryce: yeah I just noticed it today, first time I ever tried using a TTY in Intrepid [22:58] how many people switch to TTYs to perform commands when there is a terminal? [23:00] is anyone who sees this _not_ using evdev for their keyboard? [23:00] $ grep 'keyboard: xkb_rules' /var/log/Xorg.0.log [23:00] (**) AT Translated Set 2 keyboard: xkb_rules: "evdev" [23:01] I am not [23:01] bryce: would it be useful to have someone test with Hardy? [23:02] i tried this xev thing, when i typed nano in TTY1, xev said i typed nnaannoo :P [23:02] http://pastebin.osuosl.org/22218 [23:02] bdmurray: yes that would be useful; if it can be reproduced there, that would both rule out input-hotplug, and show that the fix needs backported there [23:03] kees, can you pastebin your /etc/X11/xorg.conf? [23:04] bryce: http://pastebin.osuosl.org/22219 [23:04] bryce: what good is xorg.conf? [23:04] Hamra: it'll show whether or not input-hotplug is being used [23:04] Hamra: I wanted to doublecheck that he's got driver 'kbd' rather than 'evdev' [23:04] kees, ok, so that rules out evdev as a potential culprit. [23:05] ah ok [23:05] greg-g: are you still on Hardy? [23:06] in my xorg.conf , there is a line that says Driver "kbd" , but the above command by bryce shows the exact output beneath it [23:07] my xorg.conf got copied from my hardy install BTW [23:07] FWIW, I'm on Hardy and I can't reproduce this at all [23:07] hggdh: btw for future reference, it's helpful if you can attach each file separately rather than in a tarball, since having to download and open the tar makes it a bit more difficult to review (yeah, LP sucks for posting multiple uploads, but...) [23:08] ok, also not likely to be a video driver problem... reproducers are on -intel, -radeonhd, -nvidia. Pretty broad spectrum [23:08] bryce, and I was thinking this would be easier... sorry, and will do [23:09] kees: so at this point I'm suspecting it's not X causing the problem, but something lower level [23:10] kees: it feels like something isn't properly filtering what's going on at the tty [23:10] kees: which I assume is handled at a lower level than X [23:10] bryce: I've always wanted to understand what was "below" X in this regard. tty magic is always eluded me. :P [23:11] is console-kit in between X and the tty by any chance? [23:11] yeah... it's a vague area for me too. I assume tty's are kernel-land [23:11] has anyone noticed that the longer you wait to go back to X, the less characters get leaked? [23:11] Hamra: yeah [23:12] bryce: if it's useful, I'm running the -1.2 kernel, so if it IS kernel, it's not too recent. [23:12] I don't, however, have a .26 around any more. [23:13] so we have 1 unconfirmed in Hardy but that's it? [23:14] mrooney: hrm? someone said they could do it in hardy? [23:14] nope, i said that i couldn't do it in hardy;) [23:15] hah, I *do* have a Hardy box... hold on [23:15] chrisccoulson: that's what I thought you said. :) [23:15] hggdh: i thought i was the only person running hardy for a minute? [23:15] :-) [23:15] i'm feeling left out running all this out-dated software [23:15] I forgot about this X machine... rarely use it [23:16] i must find time to upgrade this weekend [23:17] I cannot reproduce it on Hardy [23:17] that reduces the scope, good. [23:18] i've tried on intrepid in a virtual machine, but i can't reproduce it [23:18] it is quite out of dat though [23:18] i'll upgrade everything and see if it appears [23:19] is there anyway to let a hardy machine use another input method? a custom kernel maybe? [23:21] chrisccoulson, this is the single machine running X outside my laptop. All others are server, no X (but they are Hardy) [23:22] chrisccoulson: don't feel bad, I use Hardy as my main system too, I was just on my Intrepid install to test to see if a Hardy bug was fixed :) [23:22] i'm glad i'm not completely alone:) [23:23] another thing is, that we can't leak characters across TTYs, only btween TTY and X, so it must be something staying between them [23:24] I may even wait to upgrade until Dell starts shipping 8.10, then I'll probably buy an XPS M1330 [23:25] I suppose the next step would be to have someone boot a 2.6.26 kernel and see if it can be reproduced there [23:25] if not, then maybe kees should solicit a kernel dev to help look at it [23:26] * wgrant is trying to test it, but can't switch to a VT. [23:26] bryce: perhaps poke them in #ubuntu-kernel? I've got to switch gears and finish a security update publication. [23:26] ConsoleKit steals it back to X. [23:27] Ah, there we go. Finally defeated ConsoleKit. [23:27] kees: I'm in the middle of a few things myself [23:28] * wgrant ponders grabbing 2.6.26 from LP. [23:28] wgrant: that certainly makes me suspect ConsoleKit [23:28] kees: Why? It just switches the VT straight back, but I guess it could be related. [23:29] wgrant: I mean, if it has that control over the input events, perhaps it's between the kernel and X [23:29] bdmurray: I am on Intrepid now. why you ask? [23:30] greg-g: I think we got enough Hardy tests, thanks though [23:30] kees: True. [23:30] bdmurray: ah, upgrades? [23:30] bdmurray: mine went well on the desktop, there was an issue with nm-applet on the laptop which I am going to look at tonight [23:35] while I am here, is anyone aware of a cpu hungry xorg bug for Intrepid? [23:35] it loves to hang out at 15-30% for seemingly random periods of time and make rhythmbox skip in the process [23:37] mrooney: you running firefox at the same time? [23:38] greg-g: yes I do believe [23:38] I rarely don't run firefox, I can try without [23:45] kees: I've tested with 2.6.26-5 and still happens [23:47] Somebody ought try a Hardy kernel. [23:47] wgrant: hardy kernel w/ intrepid? [23:48] bdmurray: Right. [23:48] Since we know it doesn't happen on Hardy. [23:48] I can do that too [23:48] bdmurray: apparmor won't load with intrepid+.24 (but that's not important for the test). [23:49] I would test, but I don't have my normally large selection of kernels already installed because I reinstalled with alpha 5. [23:54] okay still happens with 2.6.24 and intrepid [23:56] bdmurray: okay, so not the kernel. [23:56] I'll update the bug [23:56] bdmurray: cool, looks like jdstrand tested 2.6.26 too. [23:57] he could have told somebody ;) [23:57] I added a bug comment :P [23:58] bdmurray: so, if you're in a position to do so, try installing the console-kit from hardy and rebooting with that... [23:58] I even decided to try it on a ficticious kernel (:P to self) [23:58] jdstrand: heheh [23:58] Damn. [23:58] Blaming things on the kernel is always a nice and simple way out.