vcs oralce rac 集群报错

淡然的紫色

系统环境是solaris 10、veritas vcs 5.0 、oracle rac 10g ，3个节点的集群，系统出现如下报错，css首先出现问题，然后数据库就offline了。是bug还是配置问题啊
Jul 13 15:49:52  Had[4720]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13027 Resource(cssd) - monitor procedure did not complete within the expected time.
Jul 13 15:50:01 ofs00biiacc001 AgentFramework[5980]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13027 Thread(7) Resource(cssd) - monitor procedure did not complete within the expected time.
Jul 13 15:50:01 ofs00biiacc001 Had[4720]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13027  Resource(cssd) - monitorprocedure did not complete within the expected time.
Jul 13 15:58:53 Had[4720]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13210 Agent is calling clean for resource(cssd) because 4 successive invocations of the monitor procedure did not complete within the expected time.
Jul 13 15:59:02  AgentFramework[5980]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13068 Thread(10) Resource(cssd) - cl
ean completed successfully.
Jul 13 16:00:55  Had[4720]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13077 Agent is unable to offline resource(cssd). Administrative intervention may be required.
Jul 13 17:15:54  Had[4720]: [ID 702911 daemon.notice] VCS ERROR V-16-1-13067  Agent is calling clean f
or resource(BEICENP) because the resource became OFFLINE unexpectedly, on its own.

无牙

看看网络有没有问题。

淡然的紫色

回复 2# 无牙


    网卡做了ipmp，并且日志里没有报错。

无牙

VCS的心跳网卡和Oracle的是分开的还是在一起的？

gabconfig -a   《==== 输出是什么？

fenglimian

vcs的报错是monitor脚本在连续4次（默认每次检查是60秒）检查cssd进程没有得到状态返回后认为这个资源状态可能有问题，然后就把这个资源offline，对应拥有这个资源的service group也就被offline了。

可以查一下为什么monitor不返回，有可能原因是系统非常忙，monitor没来得及返回，或者cssd真是有点问题。