[PVFS-users] Re: PVFS access hangs

[Rob Ross]

Hi Brannen,

There's very little that changes due to configure scripts, and in 
particular for a given PVFS version there are no changes to what is sent 
between client and server based on that (the protocol is fixed), so that 
shouldn't be an issue.

Do the PVFS utilities work when you have your original Xeon client 
configuration (that was hanging)?  In particular do pvfs-ping, u2p (to 
copy something onto the file system), and pvfs-ls work?

Those will help us narrow down whether it's definitely a kernel issue or 
something stranger.

Thanks!

Rob

On Fri, 9 Jul 2004, Brannen S Hough wrote:

>   I've gotten further in diagnosing my problem - I've gotten the Red Hat
> 9 system to work as a standalone (as Manager, only IONode, and Client),
> and can mount and read and write to the pvfs file system OK.  I also
> extended it by adding two other nodes (a PIII based machine and a PII
> based machine), so I have 3 IONodes running and can use the pvfs file
> system OK.
> 
>   I was running before using the PIII based system as the Manager, and
> using the Red Hat 9 system (on a Xeon) as the Client - which was
> hanging.
> 
>   How much adjusting of the source code is done by the configure script?  
> I built the binaries for each system on each system separately - is it
> possible there was some inconsistency introduced by the configure script
> due to the different architectures of the different machines I am using?