It took much work, and we still have on open PMR on the subject, but we have 
our first MSCluster up and going.  The one problem is the TDP agent on the 
cluster has ceased working.  Master, model, and msdb backup with no trouble, 
but the big production DB dies after about 12-13 minutes.  Large, unclustered 
DBs, backup on the machine regularly, and we backeup this one before defining 
the cluster.  
 
The DBA is sure he has found the solution.  When he starts the backup, this is 
the generated command:
 

BACKUP failed to complete the command USE master BACKUP DATABASE [ProdDB] TO 
VIRTUAL_DEVICE=N'TDPSQL-000017B8-0000' WITH BLOCKSIZE=512, 
MAXTRANSFERSIZE=1048576, NAME=N'full', DESCRIPTION=N'TDPSQL-000017B8-0000 (TDP 
MS SQL V2)'

He's convinced the problem is with the MAXTRANSFERSIZE.  If I would change 
that, everything would work.  But it's not a TSM parameter.  He says it doesn't 
appear in MSSQL either.  So where does this come from?  Is it really the 
culprit?  Support hasn't been very supportive so far, prefering to concentrate 
on the cluster.