paging space空间不足引起数据库宕机问题

环境：AIX 5300-04, oracle 10.2.0

现象：在Alert日志中会出现

Wed Mar 11 04:55:39 2009
Errors in file /oracle/admin/alarm/bdump/alarm_psp0_828600.trc:
ORA-27300: OS system dependent operation:fork failed with status: 12
ORA-27301: OS failure message: Not enough space
ORA-27302: failure occurred at: skgpspawn3
Wed Mar 11 04:55:40 2009
Process J000 died, see its trace file
Wed Mar 11 04:55:40 2009
kkjcre1p: unable to spawn jobq slave process
Wed Mar 11 04:55:40 2009
Errors in file /oracle/admin/alarm/bdump/alarm_cjq0_893232.trc:

oracle建议：

    As a general rule, a initial setting for the paging space is half size of memory plus 4G, with an upper limit 32G.Monitor the paging space use command 'lsps -a'.The metric %used in the output of lsps -a is typically less than 25% on a healthy system. A properly sized deployment should require very little paging space and excessive amount of swapping is an indication that the RAM on the system might be undersized.

    一般情况下，paging space的初始设置是内存大小的一半加上4G，最高不超过32G。使用命令'lsps -a'可以监控paging space 的使用情况。在一个健康的系统里，'lsps -a'输出项中的%used应该小于25%。合适的大小只产生很小的paging space，过多的swapping说明系统的RAM可能过小。

 

# lsps -a
Page Space      Physical Volume   Volume Group    Size %Used Active  Auto  Type
hd6             hdisk0            rootvg        2048MB    83   yes   yes    lv

 

    Caution:Do not undersize the paging space. If you do, the system can terminate active processes when it runs out of space. However, over-sizing the paging space  has little or no nagative impact.

    警告：不要将paging space设置过小。如果paging space设置过小，当进程运行空间溢出时会导致系统终止活动进程。然后过大的paging space很少或不会对系统产生负面影响。

 

解决办法：增加paging space 到48G，问题消失。