[GE users] occasional job failure - can't find user's home directory

reuti reuti at staff.uni-marburg.de
Wed Nov 10 18:56:35 GMT 2010


Hi,

Am 10.11.2010 um 19:05 schrieb fx <d.love at liverpool.ac.uk>:

> reuti <reuti at staff.uni-marburg.de> writes:
>
>> Am 27.10.2010 um 20:22 schrieb cjf001:
>>
>>> Ah - to avoid the dreaded "stale NFS handle" error, I guess ?
>>
>> Correct -- Reuti
>
> Is that relevant if your devices are stable through the reboot?  We
> reboot our GNU/Linux and Solaris NFS servers with no problems  
> without it
> -- things just hang until they come back.

I'm right now on the road, but I can check the details next week. The  
reason to use this hit me first with openSUSE 11.1, when they use some  
speedup with kernel NFS. If they mount /proc/fs/nfsd then you need the  
filesystem id. If you comment out this line in the nfs daemons startup  
script, you don't need to set the file system id to survive a reboot.

-- Reuti  


>
>>> That's
>>> a good idea - although in my particular case, we use NetApp  
>>> fileservers,
>>> to their export implementation is a bit a different - I'll have to
>>> see if they support the "fsid" option.
>
> -- 
> Dave Love
> Advanced Research Computing, Computing Services, University of  
> Liverpool
> AKA fx at gnu.org
>
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=294601
>
> To unsubscribe from this discussion, e-mail: [users- 
> unsubscribe at gridengine.sunsource.net].

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=294615

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list