I thought I'd give a heads up on this.  I've got an open PMR on this, and I am
waiting to hear what Tivoli/IBM has to say.

AIX 4.3.3.02
TSM server 3.7.3.1
TSM client 3.7.2.14  (applied fixtest level since we were having the lovely
unknown system error return code 195)

During the course of doing a hardware upgrade of two silver thin nodes to two
winter hawk wide nodes, a server was taken down while actively serving an NFS
mount.  This filesystem was mounted by all of the rest of the nodes in the SP
complex and the S80s.  (Pardon me if I didn't say this correctly- I am still on
the upside of the learning curve of AIX ;-)

All of these clients that had this mount failed last night and it didn't show up
in a "q ev -ex=yes".

dsmerror.log says
     07/11/00    23:35:58 PrivIncrFileSpace: Received rc=164 from
fioGetDirEntries:   /   /netmounts
     07/11/00    23:36:00 23:36:00 ANS4023E Error processing '/' : file
input/output error

dsmsched.log says
Executing scheduled command now.
07/11/00   23:33:16 --- SCHEDULEREC OBJECT BEGIN HBORACLE90_SCSP05 07/11/00
23:30:00
07/11/00   23:36:00 --- SCHEDULEREC STATUS BEGIN
07/11/00   23:36:00 Total number of objects inspected:    1,777
07/11/00   23:36:00 Total number of objects backed up:        0
07/11/00   23:36:00 Total number of objects updated:          0
07/11/00   23:36:00 Total number of objects rebound:          0
07/11/00   23:36:00 Total number of objects deleted:          0
07/11/00   23:36:00 Total number of objects expired:          0
07/11/00   23:36:00 Total number of objects failed:           0
07/11/00   23:36:00 Total number of bytes transferred:        0
07/11/00   23:36:00 Data transfer time:                    0.00 sec
07/11/00   23:36:00 Network data transfer rate:            0.00 KB/sec
07/11/00   23:36:00 Aggregate data transfer rate:          0.00 KB/sec
07/11/00   23:36:00 Objects compressed by:                    0%
07/11/00   23:36:00 Elapsed processing time:           00:02:42
***07/11/00   23:36:00 --- SCHEDULEREC STATUS END
***07/11/00   23:36:00 ANS4023E Error processing '/': file input/output error
07/11/00   23:36:00 --- SCHEDULEREC OBJECT END HBORACLE90_SCSP05 07/11/00
23:30:00
***07/11/00   23:36:00 Scheduled event 'HBORACLE90_SCSP05' completed
successfully.
07/11/00   23:36:00 Sending results for scheduled event 'HBORACLE90_SCSP05'.
07/11/00   23:36:00 Results sent to server for scheduled event
'HBORACLE90_SCSP05'.



I was trying to look up the error return code, but my errno.h only goes to 127.
Where else should I look for error return codes?  I tried searching
www.rs6000.ibm.com  but not coming up with anything useful.  I am also curious
why it returned that the event completed sucessfully-- in the past a bad or
stale NFS mount causes a failure to post, but it backs up the rest of the client
sucessfully.

The message for ANS4023E says:

ANS4023E Error processing ?filespace namepath-namefile-name?: file input/output
error
Explanation: An error was found while reading from or writing to the file.
System Action: File skipped.
User Response: Check your system to ensure that it is operating properly. For
OS/2, run CHKDSK /F for the failing drive which can be found in dsmerror.log.

Well, it did more than just skip the file.......

thanks !

lisa