How can we restore the database to a point in time and leave it in
a normal state without doing a rollforward?
Here's a sequence of db2 commands and responses that might help to
clarify the problem (this is after a restore, and the database is in a
rollforward-pending state):
---------------------------------------------
db2 => rollforward db aus_nt to end of logs and stop
SQL4970N Roll-forward recovery on database "AUS_NT" cannot reach the
specified stop point (end-of-log or point-in-time) because of missing
log file(s) on node(s) "0".
db2 =>
----------------------------
db2 =>
db2 => rollforward db aus_nt query status
Rollforward Status
Input database alias = aus_nt
Number of nodes have returned status = 1
Node number = 0
Rollforward status = DB working
Next log file to be read = S0000393.LOG
Log files processed = -
Last committed transaction = 1998-12-03-11.45.23.000000
-------------------------------------
db2 => rollforward db aus_nt to 1998-12-03-11.45.23.000000 and stop
SQL1266N Database "AUS_NT" has been rolled forward to
"2106-02-06-06.28.15.000000", which is past the specified
point-in-time.
db2 =>
-------------------
db2 => connect to aus_nt
SQL1117N A connection to or activation of database "AUS_NT" cannot
be made because of ROLL-FORWARD PENDING. SQLSTATE=57019
db2 =>
------------------
db2 => rollforward db aus_nt complete
SQL1276N Database "AUS_NT" cannot be brought out of rollforward
pending state until roll-forward has passed a point in time greater
than or equal to "1998-12-03-12.03.40.000000", because node "0"
contains information later than the specified time.
db2 =>
----------------------------------------
Thanks in advance for any input.
Larry Smith
smi...@mail.auburn.edu
Rod Merritt
Harte-Hanks Response Management