主机：HP-UX essaop1 B.11.31 U ia64 1945507590 unlimited-user license

数据库：Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production

该系统主机数据库监听服务器，启停不超过5分钟就会自动重启。业务短暂的受到影响。重启过操作系统网卡APA服务后，3个月又开始频繁重启。

xx.xx.89.3_aop数据库，监听总是重启。经过在Oracle Support官网查询，及其与一个bug相匹配。

vi /oracle/grid/11.2.0/log/essaop2/agent/crsd/orarootagent_root/orarootagent_root.log

2015-08-13 09:42:09.500: [    AGFW][10] {2:56751:2} Agent received the message: AGENT_HB[Engine] ID 12293:2614441

2015-08-13 09:42:12.618: [ default][271023]ICMP Ping from xxx.xxx.89.3 to xx.xx.89.62

2015-08-13 09:42:12.638: [ default][271023]clsicmp_pingdecode recvd other process‘s packet

2015-08-13 09:42:12.638: [ora.net1.network][271023] {0:2:9230} [check] NetworkAgent::checkLink returned false

2015-08-13 09:42:12.640: [    AGFW][10] {0:2:9230} ora.net1.network essaop2 1 state changed from: ONLINE to: OFFLINE

2015-08-13 09:42:12.640: [    AGFW][10] {0:2:9230} Switching online monitor to offline one

2015-08-13 09:42:12.641: [    AGFW][10] {0:2:9230} Started implicit monitor for [ora.net1.network essaop2 1] interval=60000 delay=60000

meatlink   Bug 16039587 。

HP UNIX下需要操作系统打个补丁：

Cause

The issue was investigated in Bug 16039587, the cause is HP-UX bug, basically the contention of address memory range lock on kernel memory causes poll(2) timeout and affects orarootagent process.

Solution

Apply OS kernel patch PHKL_42850.

找到相关补丁，进行了工单审批，到实施完成。AOP数据库监听闪断重大隐患已经解决。 

STARTED_AT                UPTIME

--------------------------------------------------------------------------------

18-SEP-2015 23:35:29      2 day(s), 10 hour(s), 2 minute(s), 54 seconds     数据库启动时间

SQL> !uptime

  9:38am  up 2 days, 11:15,  1 user,  load average: 0.03, 0.03, 0.03        主机启动时间

SQL> !lsnrctl status                                                        数据库监听启动时间

LSNRCTL for HPUX: Version 11.2.0.3.0 - Production on 21-SEP-2015 09:50:45

Copyright (c) 1991, 2011, Oracle.  All rights reserved.

Connecting to (ADDRESS=(PROTOCOL=tcp)(HOST=)(PORT=1521))

STATUS of the LISTENER

------------------------

Alias                     LISTENER

Version                   TNSLSNR for HPUX: Version 11.2.0.3.0 - Production

Start Date                18-SEP-2015 23:34:01

Uptime                    2 days 10 hr. 16 min. 44 sec

Trace Level               off

Security                  ON: Local OS Authentication

SNMP                      OFF

Listener Parameter File   /oracle/grid/11.2.0/network/admin/listener.ora

Listener Log File         /oracle/grid/base/diag/tnslsnr/essaop2/listener/alert/log.xml

Listening Endpoints Summary...

  (DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(KEY=LISTENER)))

  (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=132.35.89.3)(PORT=1521)))

  (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=132.35.89.4)(PORT=1521)))

Services Summary...

Service "+ASM" has 1 instance(s).

  Instance "+ASM2", status READY, has 1 handler(s) for this service...

Service "essaop" has 1 instance(s).

  Instance "essaop2", status READY, has 1 handler(s) for this service...

Service "essaopXDB" has 1 instance(s).

  Instance "essaop2", status READY, has 1 handler(s) for this service...

The command completed successfully

本文出自 “杨俊峰的博文” 博客，转载请与作者联系！

oracle11g监听网卡总断（二）