1 Reply Latest reply: Apr 11, 2013 3:12 AM by 1001856 RSS

    Oracle 10g RAC 2节点不定期重启

    user5696391
      Oracle 10g RAC 2节点不定期重启
      oracle数据库版本10.2.0.5,两个节点的RAC, AIX 6.1 + PowerHA 5.5
      近期2节点操作系统不定期重启,表象是心跳网络出了问题,但查找不到具体原因,
      但我发现了一个奇怪的问题
      $ crsctl get css misscount
      600
      但ocssd.log里却显示是30s
      [    CSSD]2013-04-07 15:38:51.837 [1029] >TRACE: clssnm_skgxninit: HACMP clusterware detected
      [    CSSD]2013-04-07 15:38:52.840 [1] >TRACE: clssnmNMInitialize: misscount set to (30)
      [    CSSD]2013-04-07 15:38:52.841 [1] >TRACE: clssnmStartNM: reboottime set to (3) sec
      [    CSSD]2013-04-07 15:38:52.841 [1] >TRACE: clssnmNMInitialize: Network heartbeat thresholds are: impending reconfig 15000 ms, reconfig start (misscount) 30000 ms
      按Oracle的说法,使用第三方的集群软件,misscount为600s,
      请教大家,crs的这个misscount,启动的时候,是在哪儿读取的?为什么会出现值不一致呢?