[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

abit ax8 - strange behavior



hi guys,

for something like 2 months i have amd64-based computer with abit ax8 motherboard (2 gigs of memory, 4x sata drives). system is debian testing, kernel 2.6.11-4. i could see some strange behavior (mostly) in the last days:

1) in the /var/log/messages, i can see the following
Losing some ticks... checking if CPU frequency changed.
warning: many lost ticks
Your time source seems to be instable or some driver is hogging interrupts.
this i think was happening from the beginning, however recently i observed one more error occuring:
rip _do_softirq+0x48/0xb0

i checked /proc/interrupts and the biggest "hits" are
IO-APIC-edge timer
IO-APIC-level libata
after some time, also
IO-APIC-level eth1
has very large number... timer has really high values compared to the remaining values. abovementioned error might be related to harddrives - e.g. it emerged during some rsync process and also somewhere near/after the hdd-related work during the boot

2) it might be connected to the problem above
it happened several times - the machine froze when i was writing some data on the harddrive. e.g., i wanted to copy 2 directories from a remote machine using
tar c dir1 dir2 | ssh machine "cd dir; tar xv"
this froze everytime somewhere in dir1 on the same file. then, i tried copying dir2 first and subsequent copy of dir1 proceeded ok! i could observe this for both drives connected to the same controller type VIA VT6420 (there are 2 VT6420 SATA controllers and 2 controllers denoted "unknown device 3249 (rev 50) on the MoBo). the latter don't cause any problems (it seems so).

it might be another issue - sometimes during the boot, fourth SATA drive (one of the possible troublemakers) is not recognized by BIOS. i thought that this could be caused by weak power supply (350W), however it happened even with 3 drives only... and abovementioned problems happened for both drives... i'm writing it just in case it would be useful to trace down the problems.

does anyone know what could be the problem? thank you in advance for your help.

best regards,

--
Lubos
_@_"



Reply to: