=== bjf is now known as bjf-afk === MTecknology is now known as Linux === Linux is now known as MTeck-Linux [01:27] jjohansen: fwiw, checking out from git fixed my problems 100% - thanks again [01:28] glad to hear it [01:29] i don't even know if i should log a bug for the previous issues - i could reproduce it on a completely clean environment [02:12] jjohansen: not sure if this is the appropriate forum to ask, but what's your outlook on xen support fully returning to vanilla kernels? [02:21] rackerhacker: not soon maybe 34 [02:22] yeah, i'm not feeling optimistic :/ [02:22] i noticed the fedora devs mentioned in their docs that they were hoping for .33 === Whoopie_ is now known as Whoopie [03:43] rackerhacker: not optimistic, not .33 and I doubt .34 === ripps|sleep is now known as ripps [10:48] I'm missing something in the kernel... In lucid when I try to mount anything it says 'mount: special device /dev/sda1 does not exist' that's my /boot. / is on sda2 and that mounts fine during boot. It's dying at mountall; any idea what's missing or screwed up? [11:49] smb: can we apply your extra debug messages patch to something 2.6.32 so that we continue testing this Geode issue against Lucid material? [11:50] it also seems that plymouth cannot work against a 2.6.31 kernel, anyhow. [12:14] MTeck-Linux, sounds like sda1 wasn't foudn by udev to me, check /var/log/udev and see if it has anything about sda1 [12:51] Q-FUNK, I can't say anything about the plymouth issue. I can check how well the old patches go against Lucid. Do you want the amount of debug that still crashes or the amount that suddenly makes it work? === mcasadevall is now known as NCommander [13:17] Hi, running Lucid's kernel, I no longer have a /dev/net/tun. Does anyone know why that might be? [13:17] Booting Lucid with Karmic's kernel makes it appear again [13:18] maxb: It's no longer automatically loaded. [13:18] maxb: You can wait for a new kernel (which will have it built in), or you can "modprobe tun" until then. [13:18] maxb, should be temporary. initramfs no longer creates the tun/tap device, but now the driver is gonna be built into the kernel. [13:19] What should users of, etc., openvpn clients be doing.... ah, right [13:25] smb: the amount that makes it boot :) [13:26] Q-FUNK, Ok, I try to make it happen. But it might get a bit delayed [14:09] smb: sounds good. thanks! [14:16] I have a reproducible kernel crash on lucid [14:16] how do I get a backtrace, the machine just reboots [15:58] another crash [16:42] tjaalton, i assume its not getting recorded in syslog? [16:45] apw: no, I installed linux-crashdump hoping that it would catch it [16:46] tjaalton, can you turn on kerneloops and catch it perhaps? [16:47] manjo: does it take more than installing that package? [16:47] apt-cache search kerneloops [16:48] kerneloops-daemon - kernel oops tracker [16:48] kerneloops - kernel oops tracker [16:48] kerneloops-applet - applet for the kernel oops tracker [16:48] I had the daemon already [16:49] I believe that apport hooks for kerneloops also exists [16:51] kerneloops was disabled, now running [16:53] tjaalton, kerneloops will only help if something got logged [16:53] otherwise, you will need to use linux-crashdump [16:57] they are both installed now, but it seems that linux-crashdump only installs the changelog [16:57] some packaging goof? [16:57] tjaalton, before you reboot does it give you enough time to even take a photo of the screen ? [16:58] manjo: no, it reboots itself [16:58] but of course I can't reproduce it anymore (at will) [16:58] it happened three times by just logging in, opening a terminal and then browsing some directory with tab-completion :) [17:00] tjaalton, what hw is this on ? [17:00] manjo: thinkpad x61, intel [17:00] I've got a newer libdrm/mesa though, could be related [17:08] but since linux-crashdump is essentially empty, there's no way to get a crashdump? [17:36] crashed again, but I see no dumps [17:36] l-c seems to be a metapackage.. no wonder it doesn't install anything [17:43] tjaalton, can you give us a clue as to where it crashes perhaps ? funciton name or something that flashes on the screen momentarily [17:57] manjo: the display freezes for a moment, and the next thing is the thinkpad boot logo :) [18:03] and l-c doesn't catch that. the example on wiki works === mcasadevall is now known as NCommander [19:06] ;] === asac_ is now known as asac === bjf-afk is now known as bjf [20:06] jjohansen: i've built linux-ec2 properly via git, but i always end up with my total memory inflated by ~ 9MB: http://pastie.org/private/a78gpabfjnl94cn4uciq [20:06] jjohansen: i've been digging through my .config for any option that would cause that to happen but i haven't had luck finding something [21:03] rackerhacker: are you doing i386 or amd64? [21:03] amd64 [21:04] right there was something about the mapping for the EC2 kernels [21:04] The i386 base is at 0 but amd64 isn't [21:05] I need to refresh my brain give me a minute [21:05] jjohansen: no problem ;) i appreciate the help [21:20] ericm_, ping? [22:47] NCommander, can you get éclair out & fix the camera [22:47] manjo, wrong channel? :-) [22:47] NCommander, :) yeah wrong channel but right guy [22:48] manjo, try #cyanogenmod, but I'm working on it [22:48] NCommander, they talk about GF's and such... not real andriod talk [22:48] manjo, #android-dev? [22:48] I was on it and shouted for a little bit... [22:48] hmm let me try that [22:51] NCommander, does your camera have this problem ? [22:51] manjo, well, I'm not sure what the prroblem is specifically [22:52] NCommander, solarization - the image looks like I am looking thro a film of oil ... colors are all weired [22:53] manjo, post a pic? [22:57] NCommander, email on the way [22:59] NCommander, did you get it ? [23:13] erm, this isn't normal, correct? http://pastebin.com/d4273c26d [23:21] crimsun, ah, no, that looks bad [23:22] that's stock 2.6.32-9.13-generic [23:22] purposely with all proprietary drivers deinstalled and some time after a fresh (cold) power cycle [23:23] I've reproduced it using a brand new device (assuming identical) from retail [23:23] also reproducible on i386 [23:23] (i386 is a different machine entirely) [23:23] crimsun, please file a bug, sounds like a regression [23:23] yeah, I'm trying [23:24] I can't seem to stay connected long enough to get through the LP auth process :( [23:24] can you go hard-wired? [23:25] unfortunately, no (crappy netbook) [23:26] I've already got most of it typed, so I'll just sneakernet it [23:28] any guess what I'm probably missing if all my partitions are recognized fine except for extended partitions? === bjf is now known as bjf-afk [23:30] MTeck-Linux, is your extended partition formatted ? [23:31] manjo: ya [23:32] MTeck-Linux, what does sudo fdisk -l say ? [23:32] manjo: I'm been messing aaround with the kernel [23:33] This is the first partition during boot that the kernel panics on; /dev/sda5 19 1842 14651248+ 83 Linux [23:34] /dev/sda1 or such ? [23:34] that mounts fine [23:34] ok [23:34] so your primary mounts ok [23:34] yup [23:35] you have /home on extended partition ? [23:35] what is the kernel version ? [23:35] ya [23:36] I'm using Ubuntu-2.6.32-10.14; but I've also been compiling it myself and taking some things out; that's where I'm missing something [23:36] did you modify fstab by any chance ? [23:36] ah ic [23:36] so you have a custom kernel [23:36] I could go through and enable a lot of stuff, i just thought I'd ask for an idea here [23:37] hmm not seen that failure ... iirc [23:37] not with stock kernels [23:37] one place I would check is to see if you fstab is not corrupted [23:37] it's not [23:37] ok [23:38] I only have one system; it's just something i disabled in the kernel that I need; I need to figure out what it is [23:39] you could boot into a bash shell and try to manually mount your /home [23:40] init=/bin/bash I think [23:41] it's a kernel panic so that probably won't work [23:42] do you know where it panics ? [23:42] mounting /dev/sda5 [23:42] vfs [23:44] oh... [23:44] thanks :) [23:45] now where is that :P [23:45] MTeck-Linux, did you mess with any file system config options ? [23:46] ya [23:46] like enable devfs support or something [23:46] no [23:47] but I can't find either vfs of devfs by searching for it [23:47] you can get the kernel source and look at the std config option and compare them with your modified options [23:48] the std config options should be under debian.master/config [23:48] if you have the kernel source you can look there [23:48] thanks [23:49] config.common.ubuntu should have common config options [23:51] MTeck-Linux, you could boot from live CD etc and see that your partition table is intact [23:52] I'm using the same system actually; using the default kernel until I get my own flying :) [23:53] ah so *it is* your kernel .. ok [23:59] Hi guys, sorry to bother you but bug 228302 has been triaged by myself and has been stale for a while, wondered if anyone from the team would mind having a look when you get a chance please? :) [23:59] Malone bug 228302 in baltix "[KARMIC] No DMA nor 32bits IO support" [Undecided,New] https://launchpad.net/bugs/228302