IT博客汇 | ora.crf资源异常—临时停止和禁用

ora.crf资源异常—临时停止和禁用

惜分飞发表于 2015-09-04 15:05:44

检查发现运行在win 2008平台的11.2.0.3 rac的crs的alert日志里面出现大量类似记录
CRS-2765错误

2015-09-04 00:12:10.431
[ohasd(3844)]CRS-2765:资源 'ora.crf' 已失败 (在服务器 'rac2' 上)。
2015-09-04 00:16:46.047
[ohasd(3844)]CRS-2765:资源 'ora.crf' 已失败 (在服务器 'rac2' 上)。
2015-09-04 00:21:21.479
[ohasd(3844)]CRS-2765:资源 'ora.crf' 已失败 (在服务器 'rac2' 上)。
2015-09-04 00:25:57.365
[ohasd(3844)]CRS-2765:资源 'ora.crf' 已失败 (在服务器 'rac2' 上)。

查看crfmond.log日志发现类似记录

2015-09-04 00:07:35.607: [    GPNP][19080] clsgpnp_getCachedProfileEx: [at clsgpnp.c:613] Result: (26)
 CLSGPNP_NO_PROFILE. Can't get offline GPnP service profile: local gpnpd is up and running. Use getProfile instead.

2015-09-04 00:07:35.607: [    GPNP][19080] clsgpnp_getCachedProfileEx: [at clsgpnp.c:623] Result: 
(26) CLSGPNP_NO_PROFILE. Failed to get offline GPnP service profile. 

2015-09-04 00:07:35.732: [ CRFMOND][19080]Sysmond coming up...

2015-09-04 00:07:35.732: [ CRFMOND][19080]Failed to load init file ret=1

2015-09-04 00:07:35.732: [ CRFMOND][19080]OSD error: op="scrfosm_loadInitFile" loc="read fail1" 
other="crfhome="D:\app\11.2.0\grid" and gipath="D:\app\11.2.0\grid\crf\admin\crf.ora"" dep="2"

2015-09-04 00:07:37.095: [ COMMCRS][19696]clsc_send_msg: (00000000058C98E0) NS err (12571, 12560), transport (533, 57, 0)


[  clsdmc][19676]Fail to connect (ADDRESS=(PROTOCOL=tcp)(HOST=127.0.0.1)(PORT=61022)) with status 9

[  clsdmt][19712]Listening to (ADDRESS=(PROTOCOL=tcp)(HOST=127.0.0.1)(PORT=61022))

2015-09-04 00:07:37.201: [  clsdmt][19712]PID for the Process [19672], connkey 5 

2015-09-04 00:07:37.201: [  clsdmt][19712]Creating PID [19672] file for home D:\app\11.2.0\grid 
host rac2 bin osysmond to D:\app\11.2.0\grid\osysmond\init\

2015-09-04 00:07:37.202: [  clsdmt][19712]Writing PID [19672] to the file [D:\app\11.2.0\grid\osysmond\init\rac2.pid] 

2015-09-04 00:07:37.734: [ CRFMOND][19676]mond_init: clsdms init successful

[   CLWAL][19676]clsw_Initialize: OLR initlevel [70000]

2015-09-04 00:12:10.050: [    GPNP][19676] clsgpnp_getCachedProfileEx: [at clsgpnp.c:613] Result: (26) 
CLSGPNP_NO_PROFILE. Can't get offline GPnP service profile: local gpnpd is up and running. Use getProfile instead.

2015-09-04 00:12:10.051: [    GPNP][19676] clsgpnp_getCachedProfileEx: [at clsgpnp.c:623] Result: 
(26) CLSGPNP_NO_PROFILE. Failed to get offline GPnP service profile. 

2015-09-04 00:12:10.197: [ CRFMOND][19676]Sysmond coming up...

2015-09-04 00:12:10.197: [ CRFMOND][19676]Failed to load init file ret=1

2015-09-04 00:12:10.197: [ CRFMOND][19676]OSD error: op="scrfosm_loadInitFile" loc="read fail1" 
other="crfhome="D:\app\11.2.0\grid" and gipath="D:\app\11.2.0\grid\crf\admin\crf.ora"" dep="2"

2015-09-04 00:12:11.557: [ COMMCRS][18376]clsc_send_msg: (00000000059498E0) NS err (12571, 12560), transport (533, 57, 0)

查询mos发现匹配文章Windows: CRS-2765:Resource ‘ora.crf’ has failed on server (文档 ID 1480263.1),从文中说明看是由于unpublished bug 14010695导致该问题,给出来建议是打psu到最新,但是升级psu需要停机窗口。临时想通过禁用ora.crf资源的方式来解决,在禁用该资源之前,我们先看下该资源的用途,确定是否可以禁用。

ora.crf用途
资源对应的功能是CHM.Cluster Health Monitor（以下简称CHM）是一个Oracle提供的工具，用来自动收集操作系统的资源（CPU、内存、SWAP、进程、I/O以及网络等）的使用情况。CHM会每秒收集一次数据。这些系统资源数据对于诊断集群系统的节点重启、Hang、实例驱逐(Eviction)、性能问题等是非常有帮助的。另外，用户可以使用CHM来及早发现一些系统负载高、内存异常等问题，从而避免产生更严重的问题。CHM会自动安装在下面的软件：
11.2.0.2 及更高版本的 Oracle Grid Infrastructure for Linux (不包括Linux Itanium) 、Solaris (Sparc 64 和 x86-64)
11.2.0.3 及更高版本 Oracle Grid Infrastructure for AIX 、 Windows (不包括Windows Itanium)。
根据上述描述可知ora.crf资源主要是用来收集信息的,而且在11.2.0.2之后才有,因此可以停止并禁用它

停止ora.crf资源

C:\Users\Administrator>crsctl status res -t -init
--------------------------------------------------------------------------------
NAME           TARGET  STATE        SERVER                   STATE_DETAILS
--------------------------------------------------------------------------------
Cluster Resources
--------------------------------------------------------------------------------
ora.asm
      1        ONLINE  ONLINE       rac2                     Started
ora.crf
      1        ONLINE  ONLINE       rac2
ora.crsd
      1        ONLINE  ONLINE       rac2
ora.cssd
      1        ONLINE  ONLINE       rac2
ora.cssdmonitor
      1        ONLINE  ONLINE       rac2
ora.ctssd
      1        ONLINE  ONLINE       rac2                     OBSERVER
ora.drivers.acfs
      1        ONLINE  ONLINE       rac2
ora.evmd
      1        ONLINE  ONLINE       rac2
ora.gipcd
      1        ONLINE  ONLINE       rac2
ora.gpnpd
      1        ONLINE  ONLINE       rac2
ora.mdnsd
      1        ONLINE  ONLINE       rac2

C:\Users\Administrator>crsctl stop res ora.crf -init
CRS-2673: 尝试停止 'ora.crf' (在 'rac2' 上)
CRS-2677: 成功停止 'ora.crf' (在 'rac2' 上)

C:\Users\Administrator>crsctl status res -t -init
--------------------------------------------------------------------------------
NAME           TARGET  STATE        SERVER                   STATE_DETAILS
--------------------------------------------------------------------------------
Cluster Resources
--------------------------------------------------------------------------------
ora.asm
      1        ONLINE  ONLINE       rac2                     Started
ora.crf
      1        OFFLINE OFFLINE
ora.crsd
      1        ONLINE  ONLINE       rac2
ora.cssd
      1        ONLINE  ONLINE       rac2
ora.cssdmonitor
      1        ONLINE  ONLINE       rac2
ora.ctssd
      1        ONLINE  ONLINE       rac2                     OBSERVER
ora.drivers.acfs
      1        ONLINE  ONLINE       rac2
ora.evmd
      1        ONLINE  ONLINE       rac2
ora.gipcd
      1        ONLINE  ONLINE       rac2
ora.gpnpd
      1        ONLINE  ONLINE       rac2
ora.mdnsd
      1        ONLINE  ONLINE       rac2

禁用ora.crf资源

C:\Users\Administrator>crsctl stat res ora.crf -init
NAME=ora.crf
TYPE=ora.crf.type
TARGET=OFFLINE
STATE=OFFLINE
C:\Users\Administrator>crsctl modify resource "ora.crf" -attr "AUTO_START=0" -init
C:\Users\Administrator>crsctl stat res ora.crf -init
NAME=ora.crf
TYPE=ora.crf.type
TARGET=OFFLINE
STATE=OFFLINE