OS error EEXIST was thrown when db2start was issued in a single node environment.

最新推荐文章于 2024-08-01 11:05:55 发布

转载最新推荐文章于 2024-08-01 11:05:55 发布 · 816 阅读

收录于

IBM DB2 专栏收录该内容

1 篇文章

订阅专栏

本文介绍了解决DB2数据库启动时遇到EEXIST错误的方法。此错误通常发生在DB2试图启动但发现已有共享内存存在的情况下。文章提供了具体的步骤来解决这个问题，包括强制停止DB2服务并清理IPC资源。

From：http://www-01.ibm.com/support/docview.wss?uid=swg21613865

Technote (FAQ)

Question

After db2start db2diag.log is reporting an OS error EEXIST.

2011-12-14-18.01.38.656079+120 E32782E328 LEVEL: Error (OS)
PID : 26310 TID : 46912685380880PROC : db2star2
INSTANCE: tamdbi02 NODE : 000
FUNCTION: DB2 UDB, SQO Memory Management, sqloMemCreateSingleSegment,
probe:100
CALLED : OS, -, shmget OSERR: EEXIST (17)

Cause

It means that a shared memory identifier exists for a key parameter.

Answer

DB2 is calling shmget when it is started, and that is when the EEXIST is returned by shmget indicating that shared memory exists at the time of the db2start.

That is a result of issuing db2start while the DB2 database server is already or still running.

So maybe db2 was stopped not properly before (IPC ressources left), or started twice (i.e. through autostart and manual).

To remove these messages, you can run the following:

db2stop force
ipclean -a
db2start