| === wylde_ is now known as wylde | |
| === F|shie_ is now known as F|shie | |
| [08:53] <luc4_mac> Hi! Is there any way to somehow reset all network settings? I reported bug 997767 and I noticed I can't reproduce when booting the live CD. | |
| [08:53] <ubot2`> Launchpad bug 997767 in linux "Network connection [rtl8139 / 8139too] is lost after some hours of inactivity and comes up again on user interaction" [Medium,Confirmed] https://launchpad.net/bugs/997767 | |
| === arand_ is now known as arand | |
| [16:24] <luca> Hi! I reported a bug and I discovered it seems it is not reproducible with the live CD. Do you think it is better if I reinstall or should I wait for ubuntu guys to find out the cause? | |
| [17:12] <penguin42> is this the network disappearing one? | |
| [17:13] <luca> penguin42: yes | |
| [17:14] <penguin42> luca: Is the live cd the same version you have installed or is it newer? | |
| [17:15] <njin> penguin42, hello, when clicking on a partition in nautilus, is mountall mounting it ? | |
| [17:15] <luca> penguin42: the installed version might have some newer packages as I might have upgraded some. Anyway, 12.04 for both. | |
| [17:16] <penguin42> njin: I don't think so - I've not quite figured out mountall, but I think it does all the bootup stuff | |
| [17:17] <penguin42> luca: It's going to be a tricky one to debug, so it might take a long time, so if it's annoying you it might be best to reinstall if it's not too painful; but before you do, try creating a new user and see if that user inherits the problem | |
| [17:17] <njin> ok, thanks, then is nautilus mounting the partitions ? I can easily reproduce crashes not captured monting and unmounting partitions | |
| [17:17] <penguin42> njin: I suspect it's udisks that's doign the mounting - but what is the program that you're seeing crash and when you say crash what do you mean - hang, nautilus crashing or something more subtle? | |
| [17:18] <luca> penguin42: oh... that is something simple I didn't try. I'll try that thanks! | |
| [17:19] <njin> penguin42, nautilus simply close itself, under strace the last line is 19:05:27.300576 exit_group(0) = ? | |
| [17:20] <njin> before i've got [pid 14728] 19:05:27.299574 futex(0x7fc810004684, FUTEX_WAIT_BITSET_PRIVATE, 1, {27941, 978095000}, ffffffff <unfinished ...> | |
| [17:20] <njin> [pid 14725] 19:05:27.299608 futex(0xc93e40, FUTEX_WAKE_PRIVATE, 1) = 0 | |
| [17:20] <penguin42> njin: Hmm that line doesn't say much really | |
| [17:20] <penguin42> njin: That's just it cleaning up | |
| [17:21] <penguin42> njin: This partition - where is it? Local disk? USB? network? | |
| [17:21] <njin> the partitions are on a second HD | |
| [17:21] <penguin42> njin: OK, does it end up mounted after nautilus quits? i.e. if you type mount in a terminal does it show up? | |
| [17:22] <njin> yes, mount and umount works well, just nautilus close itself | |
| [17:22] <njin> [Inferior 1 (process 14580) exited normally] | |
| [17:23] <njin> in gdb | |
| [17:23] <penguin42> hmm, that's one for the nautilus guys I think | |
| [17:24] <njin> ok, thanks, I open a report for nautilus attaching a screencast and strace and valgrind. | |
| [17:24] <penguin42> I doubt the strace/valgrind will help to be honest - and I guess they're pretty chunky | |
| [17:24] <njin> penguin42, do you think valgrind is useful or better to not attach it | |
| [17:25] <njin> ops, sorry | |
| [17:25] <njin> ok thanks again | |
| [17:25] <penguin42> njin: it doesn't sound like it's segfaulting or the like, so I doubt it's much use | |
| [17:25] <penguin42> njin: If you run nautilus from a terminal does it print any diagnostics before it exits? | |
| [17:26] <njin> tring | |
| [17:26] <njin> trying | |
| [17:26] <njin> penguin42, nothing, it simply exit | |
| [17:27] <brokenfingers> bug 1006427 Does this sound like a bad ISO error to anyone else here? Or is it just me? | |
| [17:27] <ubot2`> Launchpad bug 1006427 in Ubuntu Quantal "Live image won't boot, stuck on "stdin: Not a typewriter"" [High,New] https://launchpad.net/bugs/1006427 | |
| [17:27] <penguin42> njin: One for the nautilus guys | |
| [17:28] <penguin42> brokenfingers: Have you asked them to try changing the ribbon? | |
| [17:28] <njin> penguin42, great suggestion...ERROR:nautilus-places-sidebar.c:1885:open_selected_bookmark: assertion failed: (sidebar->go_to_after_mount_slot == NULL) | |
| [17:28] <njin> Aborted (core dumped) | |
| [17:28] <penguin42> njin: Ah - that's what they'll want to know | |
| [17:28] <brokenfingers> No, I asked them to check the MD5 | |
| [17:28] <njin> yeah | |
| [17:29] <penguin42> brokenfingers: They don't say what PPC hardware and whether they had it boot a previous version ok, they don't say if they're CD booting or anything | |
| [17:30] <brokenfingers> I'm assuming that it's a iBook, PowerBook, or G3/G4/G5 Apple Desktop | |
| [17:31] <brokenfingers> On my PBG4 the Precise PPC image has some trouble but still boots | |
| [17:31] <penguin42> brokenfingers: Probably but not definitely and which one - I mean the not a typewriter could be a red herring, it might not be why it's failing to boot | |
| [17:31] <penguin42> it could be a graphics problem | |
| [17:32] <brokenfingers> penguin42: But wouldn't it just resort to low graphics mode? | |
| [17:32] <brokenfingers> Even on my original 1998 iMac, Precise boots fine | |
| [17:33] <penguin42> brokenfingers: That's what's supposed to happen if X can't start - but anything can break; still either way it's best to find out exactly what hardware he's using; then we can see if it's PPC is broken on everything or PPC IS broken on :this set of hardware: | |
| [17:34] <brokenfingers> penguin42: Hmmm... Probably just his set | |
| [17:36] <penguin42> only one way to find out! | |
| [17:36] <brokenfingers> Oh! got it! the ubuntu qa website has it... hold on... | |
| [17:37] <brokenfingers> penguin42: His system: http://www.everymac.com/systems/apple/powerbook_g4/specs/powerbook_g4_1.67_15_hr.html | |
| [17:38] <penguin42> brokenfingers: OK, make sure it gets added to the bug report to say it's broken on that hardware - if it's only that hardware; I'd edit the title to something like [ppc: powerbook g4] Live image won't boot.... | |
| [17:39] <brokenfingers> Will do | |
| [17:39] <penguin42> what does the qa site have to say? | |
| [17:41] <brokenfingers> It just gives a link to that site, and says related bugs: only this one :) | |
| [17:41] <penguin42> oh well | |
| [17:42] <brokenfingers> I only commented 15 hrs ago, so I'll give him time to respond. | |
| [17:44] <brokenfingers> penguin42: do you know what "typewriter" refers to? I'm clueless (as always) :) | |
| [17:45] <penguin42> brokenfingers: Basically a terminal like device; I'm surprised it says typewriter, it normally says teletype | |
| [17:46] <brokenfingers> strange. is he using a version that already had it's EOL? | |
| [17:46] <penguin42> brokenfingers: A lot of the unix console/terminal goes back to the days of teletypes - basically a keyboard input and printer wired together over serial; and there are quite a few things to do with setting the rates and styles of terminal; lots of commands work with that in particular the stty command | |
| [17:47] <brokenfingers> Oh, so magbe it's a peripheral issue? | |
| [17:47] <penguin42> brokenfingers: No | |
| [17:47] <brokenfingers> oh. | |
| [17:49] <penguin42> brokenfingers: You normally see that error when you've got scripts that get run on login/startup and are normally connected to a console/terminal but are run from somewhere else for some odd reason; however it's not necessarily why the boot stops - it might just be output by something quirky in the boot scripts but be unrelated to the hang | |
| [17:50] <brokenfingers> Oh, so its waiting for something that's never going to happen | |
| [17:51] <penguin42> brokenfingers: You don't have enough information from that error to know | |
| [17:51] <brokenfingers> But I thought almost noone used stty anymore, does ubuntu use it? | |
| [17:51] <penguin42> brokenfingers: stty is still used in setup scripts behind the scenes a bit | |
| [17:51] <brokenfingers> penguin42: Can I get dependencies.txt with no-boot? | |
| [17:52] <penguin42> I don't know what you mean? | |
| [17:53] <brokenfingers> I mean, is thre a way to collect Apport info w/o a GUI or full Terminal/Console | |
| [17:54] <penguin42> no, if it's not booted you can't really collect the normal stuff | |
| [17:54] <brokenfingers> Ugh. | |
| [17:55] <brokenfingers> What a mess is there a way to get anythin helpful? | |
| [17:55] <penguin42> depends why it's unhappy | |
| [17:55] * LordOfTime peeks in | |
| [17:55] <LordOfTime> whats up now? | |
| [17:56] <brokenfingers> Well, depends how far it got booting. | |
| [17:56] <penguin42> brokenfingers: If it got far enough it might have left some logs in /var/log on the hard drive, if not then you might be able to do some form of recovery boot or pass different kernel options so you can see more debug | |
| [17:57] <penguin42> oh | |
| [17:57] <penguin42> brokenfingers: Actually, this is live cd - it won't have left any logs anywhere | |
| [17:57] <brokenfingers> Is there an alternitive? | |
| [17:58] <penguin42> brokenfingers: So the only chance is to change boot flags (if the livecd has a recovery/failsafe option) to make the kernel not hide the boot process, I think upstart also has some debug flags but I don't know them | |
| [17:58] <brokenfingers> I'll look them up...... | |
| [17:58] <penguin42> brokenfingers: Debugging boot process is a bit of a black art until it gets far enough to give you a shell where you can poke about | |
| [17:59] <brokenfingers> Yah... | |
| [18:01] <brokenfingers> This what I found: http://www.linuxplanet.com/linuxplanet/tutorials/7033/1/ | |
| [18:03] <brokenfingers> May bee he can run --verbose a boot?? | |
| === yofel_ is now known as yofel | |
| [19:12] <brokenfingers> bug 541438 Is this ready for triaging? | |
| [19:12] <ubot2`> Launchpad bug 541438 in apt ""apt-get purge" Selects another package when the requested package is not installed, but is provided by another package." [High,Confirmed] https://launchpad.net/bugs/541438 | |
| [19:37] <njin> penguin42, hello again, testing today's amd64build i've got LZMA data is corrupt, then VFS: Cannot open root device "(null)" or unknown-block(0,0): error -6 | |
| [19:37] <njin> ....., kernel panic | |
| [19:37] <njin> do you suggest to report a bug against Linux or not ? | |
| [19:38] <njin> to me is not clear that LZMA data corrupt what is | |
| [19:39] <jtaylor> lzma is a compression algorithm | |
| [19:39] <jtaylor> probably a corrupted file somewhere | |
| [19:39] <jtaylor> I suggest you check your files before filing a bug | |
| [19:40] <njin> md5 sum match, so probably is a bug in casper | |
| [20:48] <hggdh> just a comment: we run daily ISOs, well, daily. Results are always available at https://jenkins.qa.ubuntu.com/view/Quantal/view/ISO%20Testing%20Dashboard/ | |
| [23:08] <penguin42> brokenfingers: I assume you mean August 25th 201*2* | |
| [23:08] <brokenfingers> Oh- pffff YES! Sorry. | |
| [23:09] <brokenfingers> penguin42: Do you like th eidea? | |
| [23:09] <penguin42> brokenfingers: I don't know I've not looked at hugdays | |
| [23:10] <brokenfingers> Mmmmmm. They're pretty fun, from what I've heard. | |
| [23:12] <penguin42> how do bug importances work on ubiquity for installer crashes? | |
| [23:13] <brokenfingers> What do you mean? | |
| [23:14] <penguin42> well, to me a reliable installer crash is crticial - but doesn't that mean most ubiquity bugs get marked critical | |
| [23:15] <penguin42> this one has a workaround, but it involves uninstalling a package | |
| [23:15] <brokenfingers> Oh. Well for Quantal They are less important because It's not intended for installation for the public | |
| [23:15] <penguin42> ? | |
| [23:16] <brokenfingers> penguin42: what bug # | |
| [23:16] <penguin42> bug 989968 | |
| [23:16] <ubot2`> Launchpad bug 989968 in ubiquity "**CRITICAL** Unable to create '/root/cache/dconf'; dconf will not work properly." [Undecided,Confirmed] https://launchpad.net/bugs/989968 | |
| [23:16] <penguin42> I was going to mark it as critical, but I'm not sure - you'd end up with most stuff in ubuiquity ending up as critical which would be odd | |
| [23:17] <brokenfingers> OH. Its Precise. I know what you mean, better mark it as High coz it has a workaround. | |
| [23:18] <penguin42> it's not a very nice workaround | |
| [23:19] <brokenfingers> But it can be easily patched. | |
| [23:19] <brokenfingers> All need to do is remove slideshow or a script that decides | |
| [23:19] <penguin42> brokenfingers: yeh, although it's a bit tricky for a new user | |
| [23:20] <brokenfingers> Mmmmm... Then iguess it should be critical | |
| [23:20] <penguin42> heck, I've set it as high now - hohum | |
| [23:20] <penguin42> I guess that leaves critical for nukes other existing stuff on the disk :-) | |
| [23:20] <brokenfingers> Even other people agree - hence the ***CRITICAL*** :) | |
| [23:22] <brokenfingers> High is good tho | |
| [23:45] <brokenfingers> penguin42: bug 813134 Well If Installer failure is High, why is this Critical? And the asignee has had 2 month, no patch/fix/workaround. | |
| [23:45] <ubot2`> Launchpad bug 813134 in ubiquity "[DATALOSS] ubiquity overwrites luks-formatted device without warning" [Critical,Incomplete] https://launchpad.net/bugs/813134 | |
| [23:49] <penguin42> brokenfingers: Critical because it loses the data that's already there - that's even worse than not installing something extra | |
| [23:50] <brokenfingers> Ah. I see what you mean, do I just leave it alone? Let it expire? | |
| [23:50] <penguin42> brokenfingers: I thought that had been fixed though, I remember that bug - but then it makes me wonder why it got set to incomplete | |
| [23:50] <penguin42> brokenfingers: Yeh, it's got an assignee, let them deal with it | |
| [23:50] <brokenfingers> Ok. | |