[Lustre-discuss] Lustre clients failing, and cant reconnect

[Mag Gam]

By disabling  statahead_max what consequences can a client face?

# echo 0 > /proc/fs/lustre/.../statahead_max




On Fri, Sep 5, 2008 at 11:10 AM, Brian J. Murrell <Brian.Murrell at sun.com> wrote:
> On Fri, 2008-09-05 at 10:48 -0400, Jerome, Ron wrote:
>> ====================================  SERVER LOG
>> =================================================
>> Aug 31 04:02:04 oss1 syslogd 1.4.1: restart.
>> Sep  1 15:15:42 oss1 kernel: LustreError:
>> 2450:0:(ldlm_lock.c:430:__ldlm_handle2lock()) ASSERTION(lock->l_resource
>> != NULL) failed
>> Sep  1 15:15:42 oss1 kernel: LustreError:
>> 2450:0:(tracefile.c:432:libcfs_assertion_failed()) LBUG
>
> Did you notice this ASSERTION/LBUG?  All bets are off when you hit one
> of these.  A thread that LBUGs is put in to an uninteruptable sleep,
> which can certainly account for hung clients.
>
> This looks like bug 16496.
>
> b.
>
>
> _______________________________________________
> Lustre-discuss mailing list
> Lustre-discuss at lists.lustre.org
> http://lists.lustre.org/mailman/listinfo/lustre-discuss
>
>