在检查一个进程的日志时，意外发现这个问题。
错误信息如下：
/ora10g/app/admin/dbname/bdump/dbname1_j001_8567.trc Oracle DATABASE 10g Enterprise Edition Release 10.2.0.3.0 - 64bit Production WITH the Partitioning, REAL Application Clusters, OLAP AND DATA Mining options ORACLE_HOME = /ora10g/app/product/10.2.0/dbname System name: HP-UX Node name: dbname1 Release: B.11.23 Version: U Machine: ia64 Instance name: dbname1 Redo thread mounted BY this instance: 1 Oracle process NUMBER: 34 Unix process pid: 8567, image: oracle@dbname1 (J001) *** 2011-11-14 22:40:19.204 *** ACTION NAME:(GATHER_STATS_JOB) 2011-11-14 22:40:19.066 *** MODULE NAME:(DBMS_SCHEDULER) 2011-11-14 22:40:19.066 *** SERVICE NAME:(SYS$USERS) 2011-11-14 22:40:19.066 *** CLIENT ID:() 2011-11-14 22:40:19.066 *** SESSION ID:(1476.50721) 2011-11-14 22:40:19.066 WARNING:Oracle process running OUT OF OS kernel I/O resources *** 2011-11-14 22:40:55.057 WARNING:Oracle process running OUT OF OS kernel I/O resources *** 2011-11-14 22:44:30.269 WARNING:Oracle process running OUT OF OS kernel I/O resources *** 2011-11-14 22:47:40.952 WARNING:Oracle process running OUT OF OS kernel I/O resources WARNING:Oracle process running OUT OF OS kernel I/O resources |
检查了一下MOS，发现了相似的bug：
Bug 6687381 – “WARNING: Oracle process running out of OS kernel I/O resources” messages [ID 6687381.8]；
Bug 6908655 – “WARNING:Oracle process running out of OS kernel I/O resources” messages [ID 6908655.8]；
Bug 7523755 – “WARNING:Oracle process running out of OS kernel I/O resources” messages [ID 7523755.8]
这几个bug的共同点是都影响10.2.0.5一下的版本，且问题都在10.2.0.5被fixed。在11g的版本中，这几个bug的表现有所差别，第一个bug在11.1.0.7中被fixed，而另外两个需要在11.2.0.1中才被fixed。
这个问题可能导致严重的性能影响，严重的情况下甚至是数据库的崩溃，因此如果告警日志或trace文件中频繁出现这个问题，建议尽快按照补丁解决这个问题。