Hello frieds.
One more time I need your help.

I run rollforware database to end of logs and get a following error:
SQL1042C An unexpected system error occurred. SQLSTATE=58004

The db2diag catch following:

2011-01-11-09.28.54.640013-360 I447808256A573 LEVEL: Error
PID : 188512 TID : 9767 PROC : db2sysc
INSTANCE: db2inst1 NODE : 000 DB : MI_DB
APPHDL : 0-18097 APPID: *LOCAL.db2inst1.110111152129
AUTHID : DB2INST1
EDUID : 9767 EDUNAME: db2agent (MI_DB)
FUNCTION: DB2 UDB, recovery manager, sqlpForwardRecovery, probe:1230
RETCODE : ZRC=0x8B0F0002=-1961951230=SQLO_NOMEM_DBH
"No memory available in 'Database Heap'"
DIA8302C No memory available in the database heap.

2011-01-11-09.28.54.683082-360 I447808830A535 LEVEL: Severe
PID : 188512 TID : 9767 PROC : db2sysc
INSTANCE: db2inst1 NODE : 000 DB : MI_DB
APPHDL : 0-18097 APPID: *LOCAL.db2inst1.110111152129
AUTHID : DB2INST1
EDUID : 9767 EDUNAME: db2agent (MI_DB)
FUNCTION: DB2 UDB, data protection services, sqlufrol, probe:11500
MESSAGE : Roll Forward Error:
DATA #1 : Hexdump, 4 bytes
0x0700000006BF3A20 : 8B0F 0002 ....

2011-01-11-09.28.54.683274-360 E447809366A479 LEVEL: Warning
PID : 188512 TID : 9767 PROC : db2sysc
INSTANCE: db2inst1 NODE : 000 DB : MI_DB
APPHDL : 0-18097 APPID: *LOCAL.db2inst1.110111152129
AUTHID : DB2INST1
EDUID : 9767 EDUNAME: db2agent (MI_DB)
FUNCTION: DB2 UDB, data protection services, sqlufrol, probe:8180
MESSAGE : ADM1611W The rollforward recovery phase has been completed.

2011-01-11-09.28.54.683694-360 I447809846A504 LEVEL: Warning
PID : 188512 TID : 9767 PROC : db2sysc
INSTANCE: db2inst1 NODE : 000 DB : MI_DB
APPHDL : 0-18097 APPID: *LOCAL.db2inst1.110111152129
AUTHID : DB2INST1
EDUID : 9767 EDUNAME: db2agent (MI_DB)
FUNCTION: DB2 UDB, data protection services, sqlufrol, probe:8181
MESSAGE : The rollforward recovery phase completed with a return code of
-1961951230




What happend????!!

I would shut down the database instance and do an ipclean. Also, check for any messages/information at system level to see memory usage.

If you want to temporarily allocate less memory for the database, consider using DB2_OVERRIDE_BPF .

db2set DB2_OVERRIDE_BPF=<no of pages>

The database will be starting the db with the size specified above instead of the pre-configured/previously known value.

For more details of the parameter, check the documentation for your version of database.

Cheers
Sathyaram

Ok Thanks for you time.

But I'm not understand very well about a DB2_OVERRIDE_BPF.
What value I need? I have 4 GB of ram but the backup is from a machine with 16GB
What do yo thing?

I can set DB2_OVERRIDE_BPF with 2000 for example?

Try increasing dbheap db cfg and redo rollforward

I doubt that you can change a configuration parameter in the database that is in the rollforward pending state.

I had done it with success in the past. Rollforward should see the updated value when it makes a connection to the db.

Re-reading the previous updates, the target system might not have enough memory to allocate the bufferpools configured in the backup image so follow Sathyaram's suggestion first (set DB2_OVERRIDE_BPF to some low value such as 2000 and restart the instance). However, I'd have expected diag.log to contain errors related to db2 failing to allocate bp pages instead of dbheap memory and rollforward to return sql1218n instead of sql1042c. If setting DB2_OVERRIDE_BPF doesn't help, try increasing dbheap.

I restored one more time but now get an error when try connect:


SQL1084C Shared memory segments cannot be allocated. SQLSTATE=57019

I think you have a memory issue - as you have said that you source of the backup image has 16GB and restoring it to a 4GB machine.

Have you tried any of the options suggested above?

one more time....Thanks

I resolved the problem with your help.
and you can update parameters in restore or rollforward pending, you only need specify the db.