[mdlug] stale NFS file handles - achille's heal of linux?

[Ingles, Raymond]

> From: Dean Durant

> Hello, can anyone tell answer this question:   stale NFS file 
> handles, is there an equivalent problem in every OS or just 
> those that rely on NFS?     Is there a better way to share 
> files over a network?  

 If the server serving up the files goes away, you're going to have problems,
period. Unavoidable. Different fileystems have different weaknesses in this
case (and different strengths) but you're going to have problems when that
happens.

> I had a situation recently where there was a 38 - hour job 
> running on a cluster.   We needed to add a SCSI disk to the 
> file server (separate from the cluster).   The users were 
> screaming about space for their results.    We had to reboot 
> the file server to add the disk.

 Hot-swap hardware, as has been noted, will fix this. Not all that
expensive, either, particularly in relation to the kind of hardware you
need for a compute cluster.

 It would also be important to know exactly *how* the "big job" got
killed. What happened, exactly? What was the failure that led to the job
being 'killed'? (Did the job itself seg fault, or did something else kill
it? Etc. etc.)

 And, finally, priorities need to be set. If the job is that important,
adding space needs to wait - or at least to be provided in a temporary
way, like from a separate, temporary server, with a clear deadline for
when both will be shut down and the new space merged into the main NFS
server.

 Sincerely,

 Ray Ingles                                               (313) 227-2317

 "The man scarce lives who is not more credulous than he ought to be...
 The natural disposition is always to believe. It is acquired wisdom and
  experience only that teach incredulity, and they very seldom teach it
                         enough." - Adam Smith
The contents of this e-mail are intended for the named addressee only. It contains information that may be confidential. Unless you are the named addressee or an authorized designee, you may not copy or use it, or disclose it to anyone else. If you received it in error please notify us immediately and then destroy it.