We have successfully restore from TSM but when we perform rollback got error as below.

ON /db2/GQB/sapdata6 ,/db2/GQB/sapdata5 ,/db2/GQB/sapdata4 ,/db2/GQB/sapdata3 ,/db2/GQB/sapdata2 ,/db2/GQB/sapdata1 DBPATH ON /db2/GQB INTO GQB LOGTARGET /db2/GQB/log_dir/NODE0000/ WITH 2 BUFFERS BUFFER 1024
SQL2528W Warning! Restoring to an existing database that is the same as the
backup image database, but the alias name "GQB" of the existing database does
not match the alias "GPB" of backup image, and the database name "GQB" of the
existing database does not match the database name "GPB" of the backup image.
The target database will be overwritten by the backup version.
Do you want to continue ? (y/n) y
SQL2580W Warning! Restoring logs to a path which contains existing log files.
Attempting to overwrite an existing log file during restore will cause the
restore operation to fail.
Do you want to continue ? (y/n) y
DB20000I The RESTORE DATABASE command completed successfully.

************************************************** *********
** THE RESTORE PROCEDURE HAS NOW FINISHED SUCCESSFULLY **
************************************************** *********
GMU-BIQB:db2gqb 56> db2 rollforward db GQB to end of logs
SQL4970N Roll-forward recovery on database "GQB" cannot reach the specified
stop point (end-of-log or point-in-time) because of missing or corrupted log
file(s) on database partition(s) "0". Roll-forward recovery processing has
halted on log file "S0130978.LOG".
GMU-BIQB:db2gqb 57> db2 rollforward db GQB complete
SQL1276N Database "GQB" cannot be brought out of rollforward pending state
until roll-forward has passed a point in time greater than or equal to
"2011-07-08-08.18.22.000000 UTC", because node "0" contains information later
than the specified time.

Rollforward Status

Input database alias = GQB
Number of nodes have returned status = 1

Node number = 0
Rollforward status = DB working
Next log file to be read = S0130978.LOG
Log files processed = -
Last committed transaction = 2011-07-08-08.18.22.000000 UTC

GMU-BIQB:db2gqb 75> db2 "rollforward database GQB to end of logs OVERFLOW LOG PATH (/db2/GQB/log_dir)"

SQL4970N Roll-forward recovery on database "GQB" cannot reach the specified
stop point (end-of-log or point-in-time) because of missing or corrupted log
file(s) on database partition(s) "0". Roll-forward recovery processing has
halted on log file "S0130978.LOG".

Appreciated your help on this.

Looks the log file S0130978 is missing. Is S0130978.LOG present in overflow log path [/db2/GQB/log_dir] ?? Also try rollforwading to specific point in time OR to end of backup.

You can also use db2cklog command to check and see if the log file is in a good state.

GMU-BIQB:db2gqb 21> db2 rollforward db GQB to 2011-07-09-00.18.21.54
SQL1274N The database "GQB" requires roll-forward recovery and the
point-in-time must be to the end of logs.

Blazer,

S0130978.LOG exist in /db2/GQB/log_dir....

Is this the same log that was extracted during RESTORE (from your restore command -> LOGTARGET /db2/GQB/log_dir/NODE0000/)?

How far do you need to rollforward to? End of online backup?

Try rollforward to end of logs using OVERFLOW LOG PATH (/db2/GQB/log_dir/NODE0000) and NORETRIEVE.

Does not required roll-forward up to latest one but only thing roll-forward should be successful as this our quality system.

Please find output as below
db2 "rollforward db GQB stop OVERFLOW LOG PATH (/db2/GQB/log_dir)"
SQL1276N Database "GQB" cannot be brought out of rollforward pending state
until roll-forward has passed a point in time greater than or equal to
"2011-07-10-08.06.59.000000 UTC", because node "0" contains information later
than the specified time.

db2 rollforward db GQB to 2011-07-10-08.06.59.000000
SQL1274N The database "GQB" requires roll-forward recovery and the
point-in-time must be to the end of logs.

db2 rollforward db GQB to end of logs
SQL4970N Roll-forward recovery on database "GQB" cannot reach the specified
stop point (end-of-log or point-in-time) because of missing or corrupted log
file(s) on database partition(s) "0". Roll-forward recovery processing has
halted on log file "S0131143.LOG".

db2 rollforward db GQB query status[B]

Rollforward Status

Input database alias = GQB
Number of nodes have returned status = 1

Node number = 0
Rollforward status = DB working
Next log file to be read = S0131143.LOG
Log files processed = -
Last committed transaction = 2011-07-10-08.06.59.000000 UTC

But S0131143.LOG exist in /db2/GQB/log_dir/NODE0000.

It looks like you restored from another backup image. I suppose you specified "LOGTARGET /db2/GQB/log_dir/NODE0000/" on the restore command again. This is the dir where db2 will restore the logs stored in the backup image (logs you need to rollforward to end of backup).

You said "S0131143.LOG exist in /db2/GQB/log_dir/NODE0000", but specified "OVERFLOW LOG PATH (/db2/GQB/log_dir)" on the rollforward command. Why is OVERFLOW LOG PATH different from where you have S0131143.LOG (and possibly some additional logs)?


In this example, you need to issue:
db2 "rollforward db GQB to end of logs and stop OVERFLOW LOG PATH (/db2/GQB/log_dir/NODE0000) NORETRIEVE"