Hi, all.

I'm working to do a command-line restore of a database (8.2 on
Windows) to a different server.

I have run through various permutations of the process, to ensure I
understand as much as I can. Sadly, I'm not an expert (I'm better with
Oracle's rman than I am with db2 backups), and am learning as I go.

The goal here is to restore a backup into a net new database, onto a
development server. My main caution is that I don't overwrite any
existing files.

I believe I have all the required information, and the various
attempts at restoring a smaller test backup on my workstation appear
to have succeeded.

My process will be as follows:

1) create directories for non-system tablespace containers (I got
names for these from the output of a 'list tablespaces' command done
on the source db)
2) connect to the instance on the target server
3) "restore db *source_db_name* user *admin_id* using *admin_pw* from
*location_of_backup_root_folder* to c: into *new_target_db_name*
redirect"
4) enter a series of "set tablespace containers for # using (path
'*tablespace_container_folder*')" statements
5) "restore db *source_db_name* continue"
6) "get db cfg for *new_target_db_name*", review the listed config
details.

Following this process appears to have worked to this point.

As to my main concern, about overwriting existing information, db2
appears to be creating a new directory under c:\db2\node0000, called
'SQL#####' (next incremental number), and creating a dozen or so files
and a couple of directories (DB2EVENT, SQLOGDIR). I believe that as
long as I assign specific folders for the tablespace containers, I
should be fairly safe in conducting this restore.

Are there any other 'gotchas' that more experienced DBAs can point to
here? I've gone through the Data Recovery PDF in some detail, but who
knows what I may have missed in a 400+ page document.


Thanks for all input!

BD