【案例】Oracle CSSD进程HANG导致RAC节点重启原因分析笔记

时间:2016-11-04 19:20   来源:Oracle研究中心   作者:HTZ   点击:

次

天萃荷净

Oracle研究中心案例分析：运维DBA反映Oracle RAC数据库节点异常重启，通过分析RAC两节点日志得出是由CSSD进程HANG导致RAC节点重启。

下面是模拟主机OCSSD.LOG进程HANG住导致主机重启

1，环境介绍

[root@cisser2 ~]# crsctl query crs activeversion

CRS active version on the cluster is [10.2.0.5.0]

[root@cisser2 ~]# lsb_release -a

LSB Version: :core-4.0-amd64:core-4.0-ia32:core-4.0-noarch:graphics-4.0-amd64:graphics-4.0-ia32:graphics-4.0-noarch:printing-4.0-amd64:printing-4.0-ia32:printing-4.0-noarch

Distributor ID: RedHatEnterpriseServer

Description: Red Hat Enterprise Linux Server release 5.11 (Tikanga)

Release: 5.11

Codename: Tikanga

2，手动暂停OCSSD进程

[root@cisser1 ~]# ps -ef|grep d.bin

oracle 4666 4665 0 11:23 00:00:00 /oracle/app/oracle/product/10.2.0/crs_1/bin/evmd.bin

root 4746 4063 0 11:23 00:00:00 /oracle/app/oracle/product/10.2.0/crs_1/bin/crsd.bin reboot

root 5258 4874 0 11:23 00:00:00 /oracle/app/oracle/product/10.2.0/crs_1/bin/oprocd.bin run -t 1000 -m 500 -f

oracle 5348 4903 0 11:23 00:00:00 /oracle/app/oracle/product/10.2.0/crs_1/bin/ocssd.bin

root 14621 13718 0 11:44 pts/1 00:00:00 grep d.bin

[root@cisser1 ~]# kill -19 5348

30S后会生成下面的日志

3，主机messages日志

Mar 29 11:45:23 cisser1 logger: Oracle clsomon failed with fatal status 13.

这里看到状态为13，这里由于status代码不通，可能出错的原因不通，下面是常见的代码说明。

/* 10-39 are reserved for various kinds of steady state errors

* i.e. anything that comes after the group registration.

*/

clssomonretMEM = 11, /* memory allocation failure */

clssomonretCSS = 12, /* misc error in CSS layer */

clssomonretFATAL = 13, /* failure in CSS layer that should cause a reboot*/

clssomonretOCR = 14, /* misc error in OCR layer */

clssomonretOSD = 15, /* error in OSD layer used by generic code*/

/* 40-69 are reserved for various kinds of initialization errors

* i.e. anything that comes before the group registration.

*/

clssomonretCRSHOME = 40, /* CRS home is unavailable. */

clssomonretHOSTNAME = 42, /* unable to fetch hostname */

clssomonretSTDERR = 43, /* failure redirecting stderr */

clssomonretSTDOUT = 44, /* failure redirecting stdout */

clssomonretCHDIR = 45, /* failure redirecting corefile */

clssomonretARGS = 50, /* error processing arguments */

clssomonretCSSINIT = 51, /* failure initializing CSS-objects/APIs */

clssomonretCSSINIT = 51, /* failure initializing CSS-objects/APIs */

clssomonretOCRINIT = 52, /* failure initializing OCR-objects/APIs */

clssomonretOSDINIT = 53, /* error in OSD layer used by generic code*/

clssomonretMEMINIT = 54, /* unable to allocate memory during init */

clssomonretREINIT = 55, /* exceeded the CSS context reinit limit */

clssomonretINUSE = 56, /* duplicate oclsomon found */

4，ocssd于oclsomon日志

由于ocssd进程已经暂停了，所有ocssd没有任何日志信息

下面是clsomon日志信息

[root@cisser1 ~]# tail -f /oracle/app/oracle/product/10.2.0/crs_1/log/cisser1/cssd/oclsomon/oclsomon.log

2015-03-29 11:23:05.781: clsc_connect: (0x8467d70) no listener at (ADDRESS=(PROTOCOL=ipc)(KEY=OCSSD_LL_cisser1_))

2015-03-29 11:23:08.435: clsc_connect: (0x8466450) no listener at (ADDRESS=(PROTOCOL=ipc)(KEY=OCSSD_LL_cisser1_))

2015-03-29 11:23:15.684 clssomon: end of cssinit, status 0

2015-03-29 11:23:15.685 Reconfig event. (1/1/1)

2015-03-29 11:23:16.186 Reconfig event. (2/2/1)

2015-03-29 11:45:23.250 clssomon: Timeout waiting for CSS response.

5，节点2的日志信息

5.1 ocssd日志信息

[ CSSD]2015-03-29 11:44:55.852 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 50% heartbeat fatal, eviction in 29.810 seconds seedhbimpd 0

[ CSSD]2015-03-29 11:44:55.852 [633092416] >TRACE: clssnmPollingThread: node cisser1 (1) is impending reconfig, flag 1039, misstime 30190

[ CSSD]2015-03-29 11:44:55.852 [633092416] >TRACE: clssnmPollingThread: diskTimeout set to (57000)ms impending reconfig status(1)

[ CSSD]2015-03-29 11:44:56.854 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 50% heartbeat fatal, eviction in 28.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:44:58.709 [643582272] >TRACE: clssnmSendingThread: sending status msg to all nodes

[ CSSD]2015-03-29 11:44:58.709 [643582272] >TRACE: clssnmSendingThread: sent 5 status msgs to all nodes

[ CSSD]2015-03-29 11:45:02.716 [643582272] >TRACE: clssnmSendingThread: sending status msg to all nodes

[ CSSD]2015-03-29 11:45:02.716 [643582272] >TRACE: clssnmSendingThread: sent 4 status msgs to all nodes

[ CSSD]2015-03-29 11:45:07.725 [643582272] >TRACE: clssnmSendingThread: sending status msg to all nodes

[ CSSD]2015-03-29 11:45:07.725 [643582272] >TRACE: clssnmSendingThread: sent 5 status msgs to all nodes

[ CSSD]2015-03-29 11:45:10.855 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 75% heartbeat fatal, eviction in 14.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:11.857 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 75% heartbeat fatal, eviction in 13.810 seconds seedhbimpd 1

[ Oracle о CSSD]2015-03-29 11:45:12.733 [643582272] >TRACE: clssnmSendingThread: sending status msg to all nodes

[ CSSD]2015-03-29 11:45:12.733 [643582272] >TRACE: clssnmSendingThread: sent 5 status msgs to all nodes

[ CSSD]2015-03-29 11:45:16.738 [643582272] >TRACE: clssnmSendingThread: sending status msg to all nodes

[ CSSD]2015-03-29 11:45:16.738 [643582272] >TRACE: clssnmSendingThread: sent 4 status msgs to all nodes

[ CSSD]2015-03-29 11:45:19.858 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 90% heartbeat fatal, eviction in 5.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:20.744 [643582272] >TRACE: clssnmSendingThread: sending status msg to all nodes

[ CSSD]2015-03-29 11:45:20.744 [643582272] >TRACE: clssnmSendingThread: sent 4 status msgs to all nodes

[ CSSD]2015-03-29 11:45:20.859 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 90% heartbeat fatal, eviction in 4.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:21.860 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 90% heartbeat fatal, eviction in 3.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:22.673 [579823936] >TRACE: clssgmAllocateRPCIndex: allocated rpc 326 (0x2b7b1f5a1310)

[ CSSD]2015-03-29 11:45:22.673 [579823936] >TRACE: clssgmRPC: rpc 0x2b7b1f5a1310 (RPC#326) tag(146002a) sent to node 1

[ CSSD]2015-03-29 11:45:22.861 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 90% heartbeat fatal, eviction in 2.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:23.863 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 90% heartbeat fatal, eviction in 1.800 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:24.855 [633092416] >WARNING: clssnmPollingThread: node cisser1 (1) at 90% heartbeat fatal, eviction in 0.810 seconds seedhbimpd 1

[ CSSD]2015-03-29 11:45:25.667 [633092416] >TRACE: clssnmPollingThread: Eviction started for node cisser1 (1), flags 0x040f, state 3, wt4c 0 seedhbimpd 1

这里看以看到主机2在11:45:25的时候开始驱除主机1，但是主机在11:45：23分的时候就开始重启主机了，所以主机重启由于oclsomon进程导致的，而不是节点驱除导致的。

本文固定链接: http://www.htz.pw/2015/03/31/%e6%a8%a1%e6%8b%9fcssd%e8%bf%9b%e7%a8%8bhang%e5%af%bc%e8%87%b4rac%e8%8a%82%e7%82%b9%e9%87%8d%e5%90%af.html | 认真就输

--------------------------------------ORACLE-DBA----------------------------------------

最权威、专业的Oracle案例资源汇总之【案例】Oracle CSSD进程HANG导致RAC节点重启原因分析笔记