[Lustre-discuss] soft lockups on NFS server/Lustre client

[Frederik Ferner]

Hi All,

just a quick follow-up on this.

Frederik Ferner wrote:
> Robin Humble wrote:
[snip]
> I see this has been closed as duplicate of
>      https://bugzilla.redhat.com/show_bug.cgi?id=499019
> which is unfortunately not accessible to me.
> 
> On the other hand Red Hat support have just pointed me at this bug as 
> well and confirmed that it is not yet fixed in RHEL5.4. 

As you can see in this bug, Red Hat have provided a test kernel which 
we've been using on a number of machines without being able to reproduce 
the problem.

>> Lustre 1.6.6 isn't exactly recent. have you tried 1.6.7.2 on your NFS
>> exporter?

Now we've tried 1.6.7.2 on the NFS/Samba exporter and we've still seen 
the soft lockups until we upgraded to the test kernel mentioned above.

NB we've had to upgrade the Samba exporters to 1.6.7.2 anyway after 
we've turned on flock and found a LBUG there that is fixed in 1.6.7.2

So to summarize, the soft lockups are a bug in the RHEL kernel and are 
hopefully going to be fixed in a official update.

Kind regards,
Frederik
-- 
Frederik Ferner
Computer Systems Administrator		phone: +44 1235 77 8624
Diamond Light Source Ltd.		mob:   +44 7917 08 5110
(Apologies in advance for the lines below. Some bits are a legal
requirement and I have no control over them.)