由于32-bit的数据库在内存管理上面有一定限制，我们在安装部署的时候可能需要注意几个地方:

1.如果和webshere部署在同一台机器上面,需要把DB2重新作一次本地catalog。在webspshere的数据源里面配置的是
catalog后的名字，这样可以避免内存的限制，不然db2diag.log会出现以下错误:

2006-05-10-07.07.08.670350-300 I1148774C482 LEVEL: Severe (OS)
PID : 69170 TID : 7973 PROC : java
INSTANCE: db2admin NODE : 000
FUNCTION: DB2 UDB, SQO Memory Management, sqlocshr2, probe:200
CALLED : OS, -, shmat OSERR: EMFILE (24)
RETCODE : ECF=0x90000049=-1879048119=ECF_MEM_EXCEED_MAX_SEGMENTS
The number of shared memory segments attached to the calling process exceeds the system-imposed limit

2.对于32-bit的环境里面，对于user的olimits文件的修改，data和stack这两个值不要修改成-1.否则你在连接数据库的
时候，有可能会发生crash。

2006-05-09-14.55.39.664490-300 I1013480C337 LEVEL: Warning
PID : 31880 TID : 1 PROC : db2loggr (ROEEEOLD) 0
INSTANCE: db2admin NODE : 000
FUNCTION: DB2 UDB, data protection, sqlpghck, probe:1390
MESSAGE : ExtNum 113, state 401, baselsn 0000000161DB8000 nextlsn
0000000161DB83D3

2006-05-09-14.55.39.680379-300 I1013818C436 LEVEL: Warning
PID : 45492 TID : 1 PROC : db2agent (idle) 0
INSTANCE: db2admin NODE : 000 DB : ROEEEOLD
APPHDL : 0-10 APPID: *LOCAL.db2admin.060509193901
FUNCTION: DB2 UDB, data protection, sqlpterm, probe:2290
MESSAGE : Detected db2stop in progress. Will wait up to 300 seconds to
complete log archives.

3.如果是webspshere的应用，安装完后需要修改进程限制:chdev -l sys0 -a maxuproc='4096' 不然db2diag.log可能会有以下错误:

2006-05-10-11.26.25.104516-300 I350222C355 LEVEL: Error
PID : 16314 TID : 1
FUNCTION: DB2 Common, Generic Control Facility, gcf_getstate, probe:40
MESSAGE : ECF=0x9000034B Failed to get the state of the GCF fm module
CALLED : DB2 Common, Fault Monitor Facility, fmIsDB2AUTOSTARTOn
RETCODE : ECF=0x900003AC Function popen failed

2006-05-10-11.26.35.163799-300 I350578C224 LEVEL: Error
PID : 130082 TID : 1
FUNCTION: DB2 Common, Fault Monitor Facility, fmIsDB2AUTOSTARTOn, probe:40
MESSAGE : ECF=0x900003AC Function popen failed