database takes long time to process

Linux and Unix-Like UNIX and Linux Applications
1

Hi,

we currently having a issue where when we send jobs to the server for the application lawson, it is taking a very long time to complete. here are the last few lines of the database log.

 
2012-09-18-10.35.55.707279-240 E244403536A576     LEVEL: Warning
PID     : 950492               TID  : 27243       PROC : db2sysc 0
INSTANCE: <INSTANCE>               NODE : 000         DB   : <INSTANCE>D1
APPHDL  : 0-506                APPID: GA65441A.ED3B.120918135356
AUTHID  : <user>
EDUID   : 27243                EDUNAME: db2agent (<INSTANCE>D1) 0
FUNCTION: DB2 UDB, lock manager, sqlpRefillAgentLRBCache, probe:1000
DATA #1 : <preformatted>
Locklist has synchronously grown into overflow memory to avoid lock escalation.
Previous locklist size: 2464
Current locklist size: 2592
2012-09-18-10.35.56.563842-240 E244404113A563     LEVEL: Warning
PID     : 950492               TID  : 27243       PROC : db2sysc 0
INSTANCE: <INSTANCE>               NODE : 000         DB   : <INSTANCE>D1
APPHDL  : 0-506                APPID: GA65441A.ED3B.120918135356
AUTHID  : <USER>
EDUID   : 27243                EDUNAME: db2agent (<INSTANCE>D1) 0
FUNCTION: DB2 UDB, lock manager, sqlpCheckAndDecreaseLocklistOverflowUsag, probe:80
DATA #1 : <preformatted>
Locklist overflow reduced.
Previous locklist size: 4160   New locklist size: 2336
Pages still in overflow: 0
2012-09-18-10.47.25.221334-240 I244404677A513     LEVEL: Info
PID     : 950492               TID  : 119764      PROC : db2sysc 0
INSTANCE: <INSTANCE>               NODE : 000         DB   : <INSTANCE>D1
APPHDL  : 0-11                 APPID: *LOCAL.DB2.120918060823
AUTHID  : <INSTANCE>
EDUID   : 119764               EDUNAME: db2stmm (<INSTANCE>D1) 0
FUNCTION: DB2 UDB, buffer pool services, sqlbAlterBufferPoolAct, probe:90
MESSAGE : Altering bufferpool "BPTAB_4K" From: "14608" <automatic> To: "21912"
          <automatic>
2012-09-18-10.53.52.459546-240 E244405191A576     LEVEL: Warning
PID     : 950492               TID  : 27243       PROC : db2sysc 0
INSTANCE: <INSTANCE>               NODE : 000         DB   : <INSTANCE>D1
APPHDL  : 0-506                APPID: GA65441A.ED3B.120918135356
AUTHID  : <USER>
EDUID   : 27243                EDUNAME: db2agent (<INSTANCE>D1) 0
FUNCTION: DB2 UDB, lock manager, sqlpRefillAgentLRBCache, probe:1000
DATA #1 : <preformatted>
Locklist has synchronously grown into overflow memory to avoid lock escalation.
Previous locklist size: 2464
Current locklist size: 2592
2012-09-18-10.53.53.057578-240 E244405768A563     LEVEL: Warning
PID     : 950492               TID  : 27243       PROC : db2sysc 0
INSTANCE: <INSTANCE>               NODE : 000         DB   : <INSTANCE>D1
APPHDL  : 0-506                APPID: GA65441A.ED3B.120918135356
AUTHID  : <USER>
EDUID   : 27243                EDUNAME: db2agent (<INSTANCE>D1) 0
FUNCTION: DB2 UDB, lock manager, sqlpCheckAndDecreaseLocklistOverflowUsag, probe:80
DATA #1 : <preformatted>
Locklist overflow reduced.
Previous locklist size: 4256   New locklist size: 2336
Pages still in overflow: 0

does anything here look odd?

2

I do not know db2 that well, but you appear to have excessive lock contention.
This means a lot of processes are trying to select/insert/execute (pick one) a single database object. It probably means one process is sitting inactive on the object for a long time.

Users who login, then start DB work, but take a break can cause this. Their process can have a lock which does not get released until they complete the transaction they started. We have users who do not log off overnight and leave screens open. We kill off processes like that to prevent lock contention.