Re: Kernel Hangs, EIP is at scsi_decide_disposition

[Thomas Backlund <tmb@mandriva.org>]

Richard Mollel wrote:
> Our mandriva boxes are randomly crushing after
> migration to 2006.0 download version. The same
> hardware has had mandriva 10.0 withuptimes of over 230
> days....but nowadays it is hardly a week before a
> crash for over 6 nodes with latest kernels.

The correct place to report this is:

http://qa.mandriva.com/


> I captured one of our lockups and am attaching the
> output of ksymoops for further interpretation by mroe
> experienced eyes when it comes to kernel internals.
>
> Here is the actual error, and attached, the ksymoops
> interpretation of it.
>
> Unable to handle kernel NULL pointer dereference at
> virtual address 000002a9
>  printing eip:
> f88642bd
> *pde = 3702b001
> Oops: 0000 [#1]
> SMP
> Modules linked in: sg nfsd exportfs raw md5 ipv6 nfs
> lockd nfs_acl sunrpc parport_pc lp parport e1000
> af_packet floppy video thermal tc1100_wmi processor
> fan container button battery ac ipt_LOG ipt_REJECT
> ipt_state ip_conntrack iptable_filter ip_tables ide_cd
> loop hw_random 3w_9xxx tsdev sr_mod ehci_hcd uhci_hcd
> usbcore evdev reiserfs sd_mod ahci ata_piix libata
> scsi_mod
> CPU:    0
> EIP:    0060:[<f88642bd>]    Tainted: G   M  VLI
> EFLAGS: 00010282   (2.6.12-22mdksmp)

Can you reproduce with an untainted kernel ??

This is what you also need to report to the above mentioned address...

--
Thomas

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/