From: ADSM: Dist Stor Manager on behalf of John Bremer
>run a select * from backups with the required node_name, filespace_name,
>etc to find the object id, then delete object 0 xxxxxxxxx

...and guess what happens if you do this? If everything is copacetic, it deletes the 
object.
 
And if everything is *not* copacetic? It will corrupt the database and down you go If 
you are lucky, it will go down immediately; if you are not, it may not go down for a 
day or three. And when you call Tivoli support and tell them what you did, they'll 
tell you, "Gee, 'delete object' is an unsupported command, to be used only at the 
direction of Tivoli support. Guess you'll have to restore your database from your last 
good backup. Thanks for playing!" If it's been two or three days since the last good 
db backup you've done, your TSM system is good and truly hosed.
 
Do yourself a favor, Calvin. *Do not* do this. As Mr. Sims commented in an earlier 
posting, stay within the bounds of policies established by your IT infrastructure. 
Deleting a single file pointer from the TSM database isn't worth the risk. Why do you 
want to do this anyway?
 
--
Mark Stapleton

Reply via email to