两节点RAC不断重启

parknkjun

说明：数据库在1月12号不断重启，数据库是11.2.0.4 64位，os是 AIX 64位

Liu Maclean(刘相兵

node1

2015-01-12 09:24:27.628: [    CSSD][2063]clssnmvDiskPing: Writing with status 0x3, timestamp 1421025867/706585554
2015-01-12 09:24:28.082: [    CSSD][2577]clssnmvDHBValidateNcopy: node 2, wms2, has a disk HB, but no network HB, DHB has rcfg 315891916, wrtcnt, 19771779, L
ATS 706586008, lastSeqNo 19771778, uniqueness 1421024974, timestamp 1421025909/706217201



node2:

2015-01-12 09:24:27.628: [    CSSD][2063]clssnmvDiskPing: Writing with status 0x3, timestamp 1421025867/706585554
2015-01-12 09:24:28.082: [    CSSD][2577]clssnmvDHBValidateNcopy: node 2, wms2, has a disk HB, but no network HB, DHB has rcfg 315891916, wrtcnt, 19771779, L
ATS 706586008, lastSeqNo 19771778, uniqueness 1421024974, timestamp 1421025909/706217201


确定私有心跳一致没问题？

诊断此问题 需要osw及ping 日志

parknkjun

Liu Maclean(刘相兵 发表于 2015-1-14 19:14
node1

2015-01-12 09:24:27.628: [    CSSD][2063]clssnmvDiskPing: Writing with status 0x3, timestamp ...

我分析下来，也确定是心跳问题，如Number of processor cores in the system is 28
Private Interface 'en0' configured from GPnP for use as a private interconnect.
  [name='en0', type=1, ip=169.254.23.59, mac=e4-1f-13-d8-5c-fc, net=169.254.0.0/16, mask=255.255.0.0, use=haip:cluster_interconnect/62]
这里显示私有网卡为en0，但是hosts里记录的私有网卡是en4,且IP为10网段，客户现在发映不是这个问题，因为没有做任何改动，数据库现在好了，很奇怪！