[02:16] <jj-afk> back on later
[10:41] <Ntemis> hi
[10:41] <Ntemis> am facing a usb problem and i cant find a solution
[10:42] <Ntemis> my usb that has the server os on it gets disconnected randomly
[10:42] <Ntemis> i can post a log if anyone here can share a thought or 2
[10:44] <Krunch> pastbin
[10:45] <Ntemis> okey thanks
[10:47] <Ntemis> http://pastebin.de/15546
[10:50] <Ntemis> Feb 23 06:44:35 UBUNTUNAS kernel: [  512.191039] NFSD: starting 90-second grace period
[10:50] <Ntemis> Feb 23 06:44:51 UBUNTUNAS kernel: [  528.790084] usb 4-1: USB disconnect, address 2
[10:51] <Ntemis> Feb 23 15:10:02 UBUNTUNAS kernel: [  131.011075] NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
[10:51] <Ntemis> Feb 23 15:10:02 UBUNTUNAS kernel: [  131.012212] NFSD: starting 90-second grace period
[10:51] <Ntemis> Feb 23 15:10:33 UBUNTUNAS kernel: [  162.375496] usb 4-1: input irq status -75  received
[10:51] <Ntemis> Feb 23 15:10:34 UBUNTUNAS kernel: [  162.540077] usb 4-1: USB disconnect, address 2
[10:58] <Ntemis> Feb 20 16:25:46 UBUNTUNAS kernel: [  946.790077] usb 4-1: USB disconnect, address 2
[11:01] <Ntemis> any ideas?
[11:02] <Krunch> looking
[11:02] <Ntemis> thanks
[11:02] <Ntemis> if you need any other log please tell me
[11:11] <Krunch> this is your keyboard being disconnected
[11:11] <Krunch> the -75 error looks like the keyboard sending invalid usb message (EOVERFLOW)
[11:12]  * Krunch double check 2.6.37
[11:15] <Ntemis> keyboard was disconnected by me after i press enter
[11:16] <Krunch> ok, so what is the actual problem?
[11:16] <Ntemis> grub menu was not auto load the kernel so i had to connect keyboard so os would load
[11:16] <Krunch> well, the logs show the keyboard disconnecting after linux has booted
[11:16] <Ntemis> i loose connection with the server randomly
[11:16] <Krunch> do you have console access?
[11:16] <Ntemis> nope
[11:16] <Ntemis> nothing
[11:16] <Ntemis> no console no webgui access
[11:17] <Krunch> i mean do you normally have console access?
[11:17] <Ntemis> yes of course :)
[11:17] <Krunch> i mean like serial, ilo, drac,... (not ssh)
[11:17] <Ntemis> serial is disabled in bios
[11:17] <Ntemis> only ssh
[11:18] <Krunch> do you have root on another box on the same LAN?
[11:18] <Ntemis> didnt understand this
[11:18] <Ntemis> what do you mean
[11:20] <Krunch> your box is crahing or hanging, local logs are probably not reliable, there are ways around that to collect relevant, reliable data but it usually requires either console access or administrative access to a system on the same network
[11:20] <Ntemis> i have an openwrt router with 4 gbit lan and the server is connected to one of them
[11:20] <Ntemis> i have 3 pcs that can do that
[11:21] <Krunch> that said, there are still some interesting things in these logs (the hung task backtraces)
[11:23] <Krunch> i see 13 reboots between the 20th at 07:54 and the 26th at 10:41, can you tell which ones if any were normal scheduled reboots and which ones were result of the problem?
[11:23] <Ntemis> is there anything i can do from ssh to collect more info for you?
[11:23] <Ntemis> no reboots at all
[11:23] <Ntemis> i force shutdown
[11:24] <Krunch> ok
[11:24] <Ntemis> with hardware button
[11:24] <Ntemis> hmmm
[11:24] <Krunch> i don't think there is much more we can get when the system is running fine
[11:24] <Ntemis> ok let me clear this out
[11:24] <Ntemis> i have installed kernels from the mainline to test the hanging
[11:25] <Ntemis> the only kernel that holded was the 2.6.37.1
[11:25] <Ntemis> 2.6.37 was the most hangfull
[11:25] <Ntemis> 2.6.37.2 is so so
[11:25] <Krunch> you never saw the problem with 2.6.37.1 ?
[11:26] <Ntemis> nope
[11:26] <Ntemis> rock solid but only for 3-4 days
[11:26] <Ntemis> after those i installed 2.6.37.2 and removed 2.6.37.1
[11:27] <Ntemis> i have a small space and i need the mb
[11:27] <Ntemis> os is only on a 4gb usb flash drive
[11:28] <Krunch> ok, i would suggest to do the following
[11:28] <Krunch> setup kdump
[11:29] <Krunch> enable hung_task_panic
[11:29] <Krunch> enable nmi watchdog
[11:30] <Krunch> next time the problem happens, it is likely to cause a proper kernel panic and generate a vmcore (wherever you told kdump to put it)
[11:30] <Krunch> the vmcore will contain more informations about the cause of the crash
[11:30] <Krunch> right now there is not enough to determine what's going on
[11:30] <Ntemis> ok thanks
[11:30] <Krunch> although this is probably related: http://paste.debian.net/108901/
[11:30] <Ntemis> brb
[11:31] <Krunch> but that doesn't really tell us what is hanging
[11:31] <Ntemis> wife :)\
[11:33] <Krunch> (well, it's likely to be a storage problem but that can still be many things)
[12:03] <Ntemis> you still here?
[12:05] <Krunch> yes
[12:09] <Ntemis> klutch do you thing if i disable ipv6 will improve anything
[12:10] <Ntemis> Krunch: do you thing if i disable ipv6 will improve anything
[12:11] <Krunch> i don't see how this would change anything
[12:11] <Krunch> changing filesystem might help
[14:18] <PhoenixSTF> Hello guys, how do I make m HDD slow its transmition speed?
[14:44] <PhoenixSTF> apw, yout here m8?
[14:53] <Ntemis> hi
[14:53] <Ntemis> my kernel log reports i  must run the e2fsck
[14:53] <Ntemis> how i do that?
[14:53] <Ntemis> it read the hdd's must me unmounted
[14:54] <Ntemis> is there a command to check the hdds after a reboot?
[15:06] <charlie-tca> Ntemis: either boot using recovery menu or a live cd
[15:13] <Ntemis>          VVVVBFGVG G GGHGGHHHJK;
[15:21] <charlie-tca> Ntemis: was that a stroke or a mistake?
[15:44] <hallyn> the 2.6.38 kernels seem to make my vaio laptop fan always race, whereas 2.6.37-12 is nice and quiet...
[15:45] <Krunch> do you have the kernel logs at boot for each version?
[15:48] <hallyn> hm, i think so
[15:49] <hallyn> anything in particular i'd look for?
[15:50] <hallyn> acpi msgs look the same
[15:51] <hallyn> maybe i just need to tweak /sys/bus/acpi/drivers myself?
[15:51] <Krunch> that's the first thing i would have looked at but someone who actually knows something about acpi might find something by looking at other non obvious difference
[15:52] <hallyn> thx.  i'll take another closer look later, and file a bug if it persists
[18:46] <jj-afk> hallyn: what is the load like
[18:47] <jj-afk> yes, I am think acpi drivers too but I was curious if there are apparent load changes as well