[00:05] cwillu: kernel oopses when it tries to allocate memory [00:05] what are you doing when that happens? [00:15] johanbr, I'll come back to the machine and it's locked up (monitors are in stand-by, can't ssh in, etc) [00:15] so my best answer is "nothing" :p [00:15] what was it doing at the time? [00:15] happens every couple days [00:15] again, as far as I know, there is nothing out of the ordinary running at that time [00:17] basically idle? [00:17] yep [00:17] I _do_ have a daily backup happening, but I turned it off for a couple days, still experienced this hang [00:20] that shouldn't happen [00:21] oopsing when allocating memory makes me think "RAM problem" [00:21] have you run memtest? [00:21] I ran memtest on it for about 24 hours a week or two ago [00:21] no errors [00:22] 64bit cpu running a 32bit install, in case that's relevant [00:22] how much ram? [00:22] 2 gigs [00:24] hmm... no immediate guess [00:24] do you have other oopses captured? [00:27] this was the first netconsole capture I've done on it, but it lives long enough to write to the log files [00:27] May 21 13:02:43 ralph-desktop kernel: [423191.559094] BUG: unable to handle kernel NULL pointer dereference at 00000001 [00:29] looks like I've got a couple dozen in /var/log/kern.log.*.gz [00:29] q [00:30] what's the line corresponding to [129075.799432] IP: [] __kmalloc+0x6f/0x1b0 ? [00:30] this started after upgrading to jaunty, the issue remains under the 2.6.30 kernels as well [00:30] one sec [00:30] May 21 13:02:43 ralph-desktop kernel: [423191.559103] IP: [] kmem_cache_alloc+0x49/0xb0 [00:31] that one's off 2.6.28-12-generic [00:31] okay [00:31] actually, give me one sec [00:31] I just lied to you :p [00:31] oh, nevermind, I didn't lie [00:31] :) [00:31] seems like it consistently crashes when trying to allocate memory [00:32] May 29 00:32:59 ralph-desktop kernel: [567299.666278] IP: [] __kmalloc+0x67/0x180 is the one that actually matches the second one I linked above [00:32] alright... same thing, basically [00:32] but kmem_cache_alloc shows up on some of them too [00:33] will jaunty boot under a 8.10 kernel? [00:33] that should work [00:33] but you said it showed up under 2,6,28? [00:33] yes [00:34] oh, never mind... [00:34] I'm thinking "regression added in 2.6.28 that hasn't been noticed consistently yet", but my thoughts are suspect :p [00:34] it didn't happen in intrepid? [00:34] not that I know of [00:35] have you run intrepid since the bug first started appearing? [00:35] no [00:35] a bug in the kernel memory allocation would've been quickly noticed [00:35] so I still suspect some kind of hardware problem [00:35] I updated to jaunty because of the improved wacom handling (was basically unusable in intrepid due to how intrepid's xserver handled hotplugging), a day or two later he started complaining about lockups [00:36] does the machine have swap? [00:36] yes [00:36] could also be a disk problem [00:37] if you have some unpartitioned space (or an expendable partition) you could try using a different swap partition [00:37] * cwillu turns off swap and warns him of potentially odd behaviour :p [00:38] you can also do that :) [00:38] how often does the bug strike? [00:42] once a day, once every couple days [00:43] alright [00:43] at the moment, I don't have any other ideas [00:43] try running without swap for a while and see if that changes anything [00:45] will do [05:00] is the PAE x86 server kernel getting the axe? [09:30] hey guys im trying to find the docs on my device....i did lsusb and found WinChipHead but when i google it the only sites that appear are in some asian language.....is there any other way i can figure out what i need to write the driver [15:59] log_buf_len=n Sets the size of the printk ring buffer, in bytes. Format: { n | nk | nM } n must be a power of two. The default size is set in the kernel config file. http://www.kernel.org/doc/Documentation/kernel-parameters.txt [16:00] Um... I don't see it in juser@dell30:~$ grep -i log_buf /boot/config-2.6.28-11-generic [16:00] CONFIG_LOG_BUF_SHIFT=17 [16:01] "The default size is set in the kernel config file." sounds odd. [17:26] I need an 8.10 kernel config, or a 9.10 kernel config [17:54] hi.. [17:54] i am using ubuntu jaunty.. [17:54] any latest RT kernel ? [17:56] hi.. anybody pls.. [17:57] i need RT kernel for jaunty.. [17:57] the current RT kernel hangs my system.. :( [18:02] hey.. [18:02] anybody here? [22:04] mohan_... should stick around longer next time :( [22:40] https://bugs.launchpad.net/ubuntu/+source/virtualbox-ose/+bug/347487 [22:40] Malone bug 347487 in virtualbox-ose "starting virtual machine in virtualbox-ose freezes system" [Medium,Confirmed] [22:46] hello, is this a good place for questions about upgrading my kernel? [22:52] I'm going to upgrade to 2.6.30, downloaded from kernel.ubuntu.com, but the package description warns me that I may want to install the -generic metapackage instead, and I cannot find it. Is it really necessary? [22:56] arielCo, nothing is REALLY necessary [22:56] arielCo, https://wiki.ubuntu.com/KernelMainlineBuilds [22:56] @darthanubis: going [22:57] ?? [22:58] Okay, that's what I did the last time I tried (2.6.29). That time, the b44 driver did not autoload on boot and I had to type 'modprobe b44' every time I booted. [23:16] that's unrelated to not installing the -generic package [23:22] Johan, thanks. Do you know where can I find info on autoloading / hw detection? I'm about to upgrade via dpkg, and I want to understand the issue so I can transmit simple instructions to anyone doing the same :) === lifeless is now known as all === all is now known as lifeless