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

Bug#760393: Kernel 3.2 hang on a HP ProLiant ML350 G4p, Smart Array 641 controller



Package: linux-image-3.2.0-4-686-pae
Version: 3.2.60-1+deb7u3


I've just upgrade a server as in subject from squeeze to wheezy: the
server run squeeze for years, without troubles or unexpected
reboots/failures.

Upgrade went well, but after rebooting to the new kernel system became
instable and hang ''randomly'' after some hours or some days (2-3 max).

Clearly, the server is in production and on a remote location. ;(

I've managed to reboot it on the old 2.6 kernel, and the server is now
stable as before, so i suppose is a kernel trouble.

Some sparse info:

1) server hang because /var ''desappear''; i've managed to go on
 console, and what i see is only some gereric error like:

	task XXXX:YYYY blocked for more than 120 seconds.
	"echo 0 > /proc/sgs/kernel/hung_task_timeout_secs" disables this message.

no PANIC or something like that.

2) /var is a XFS filesystem.

3) because my log is on /var, i've no more info to provide...


Attached the boot log of the two kernels; the only strage thing i note
is:
 Aug 12 13:59:03 rita kernel: [    1.231468] HP CISS Driver (v 3.6.26)
 Aug 12 13:59:03 rita kernel: [    1.231583] cciss 0000:09:02.0: PCI IRQ 72 -> rerouted to legacy IRQ 16
 Aug 12 13:59:03 rita kernel: [    1.231665] cciss 0000:09:02.0: Controller reports max supported commands of 0, an obvious lie. Using 16.  Ensure that firmware is up to date.

But i don't know if is related.


Say me if more info are needed. Thanks.

Attachment: rita-2.6-good.klog.gz
Description: Binary data

Attachment: rita-3.2-bad.klog.gz
Description: Binary data


Reply to: