[OpenAFS] Problem with mandatory profiles on AFS.

Jeffrey Altman jaltman@secure-endpoints.com
Tue, 19 Aug 2008 11:08:55 -0400


Anders Magnusson wrote:
> Jeffrey Altman wrote:
>> Debugging techniques are described in the Release Notes.
>>   
> Yes, I've read them, but I didn't really understand how to parse all
> information
> that I could get out from the system :-)

Parsing the output requires an understanding of the source
code and of Windows internals.

Support is available if you cannot help yourself

  http://www.secure-endpoints.com/support/openafs.html

>> If the client comes to a complete halt it is likely a deadlock.
>> A number of deadlocks have been fixed in the 1.5.51 and 1.5.52
>> releases that have been present for quite some time.
>>   
> No, it do not look like a deadlock, everything else continues and after the
> timeout (error popup) everything is back to normal.
> 
> We have tried with 1.5.52 also, but no difference.

Without access to the information that would be provided by obtaining
OpenAFS trace output combined with SysInternals Process Monitor logging
it is not possible to provide much in the way of advice.

SMB timeouts occur when a request made to the file server fails to
respond in 45 seconds.  It is crucial that file server bottlenecks
be avoided.  Running the latest AFS servers, 1.4.7, will help prevent a
single client using a large number of worker threads on a single server.

Process Monitor output will show you which SMB client requests are
failing and how long they are taking.  The OpenAFS trace output will
show you why they are failing.

The Rx Hard Dead, Connection, Idle, and other Timeout values can
be adjusted on the client via the registry.   The options are
described in Appendix A of the release notes.