We hit the same problem a few months ago, and were told that the message is
actually misleading. Any file system, either local or remote, can be hit by a
timeout condition, raised for many potential reason.

A couple of possible solutions:

Modify the client option nfstimeout to something larger,

rename the program /usr/lpp/adsm/bin/dsmstat to something different. This one
is the one that will raise the timeout status, if I understand correctly. Not
having it around is armless, as far as we have seen.

Info on both options is available in the clients guide.

BTW, I understand that IBM is presently working on solving/bypassing this
problem.

--
Regards
Regards

+-------------------+----------------------------------+---------------------+
| Francis Dequenne  | Systems Section                  |      /~~\  /~~\     |
| ECMWF             | e-mail: fdequenne AT ecmwf DOT int      |     /    \/    
\    |
| Shinfield Park    | Tel:    (+44 118) 9499361        |   ECMWF             |
| Reading           | Fax:    (+44 118) 9869450        |   ECMWF             |
| Berkshire RG2 9AX | Telex:  (+44 118) 9847908        |     \    /\    /    |
| United Kingdom    |                                  |      \__/  \__/     |
+-------------------+----------------------------------+---------------------+