现场人员反映一套库启动失败，集群资源启动失败，发截图和日志过来分析

分析日志后发现： ASM没有启动 OCR等不能启动

问题1：

2018-05-08 11:59:39.980: [? OCRASM][1639659296]proprasmo: Error in open/create file in dg [OCRDG]

[? OCRASM][1639659296]SLOS : SLOS: cat=7, opn=kgfoAl06, dep=15077, loc=kgfokge

2018-05-08 11:59:39.980: [? OCRASM][1639659296]ASM Error Stack : ORA-15077: could not locate ASM instance serving a required diskgroup

2018-05-08 11:59:39.981: [? OCRASM][1639659296]proprasmo: kgfoCheckMount returned [7]

2018-05-08 11:59:39.981: [? OCRASM][1639659296]proprasmo: The ASM instance is down

2018-05-08 11:59:39.981: [? OCRRAW][1639659296]proprioo: Failed to open [+OCRDG]. Returned proprasmo() with [26]. Marking location as UNAVAILABLE.

?

第二个报错提示： 大概心跳网络有问题

问题2：

2018-05-08 13:06:36.097: [??? CSSD][3379095296]clssnmvDHBValidateNcopy: node 1, rac1, has a disk HB, but no network HB, DHB has rcfg 421249059, wrtcnt, 94952864, LATS 861554, lastSeqNo 94952861, uniqueness 1525754229, timestamp 1525755985/867374

ping验证私有网络没有问题

?

问题1很明显，Oracle rac asm方式，OCR存放在asm里面，启动之前确保asm实例启动

由于服务器重启后没把asm实例带起来。我们必须手工启动asm实例

sqlplus /? sysasm

startup

启动asm实例后我们启动集群相关资源

[[email protected] ~]# crsctl start has

[[email protected] ~]# crsctl start crs

[[email protected] ~]# crsctl check crs

CRS-4638: Oracle High Availability Services is online

CRS-4537: Cluster Ready Services is online

CRS-4529: Cluster Synchronization Services is online

CRS-4533: Event Manager is online

启动实例资源等

crsctl start cluster –all

验证登陆节点1登陆正常

?

处理第二个节点

?

?

?

?

?

?

总结：现场人员rac启动顺序不熟悉，Oracle grid root用户命令混用