HAIP异常导致RAC节点无法启动的解决方案.docx

HAIP异常，导致RAC节点无法启动的解决方案一个网友咨询一个问题，他的11.2.0.2 RAC(for Aix),没有安装任何patch或PSU。其中一个节点重启之后无法正常启动，查看ocssd日志如下：2014-08-09 14:21:46.094: CSSD5414clssnmSendingThread: sent 4 join msgs to all nodes2014-08-09 14:21:46.421: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 0s2014-08-09 14:21:47.042: CSSD4129clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958157, LATS 1518247992, lastSeqNo 255958154, uniqueness 1406064021, timestamp 1407565306/15017580722014-08-09 14:21:47.051: CSSD3358clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958158, LATS 1518248002, lastSeqNo 255958155, uniqueness 1406064021, timestamp 1407565306/15017581902014-08-09 14:21:47.421: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 02014-08-09 14:21:48.042: CSSD4129clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958160, LATS 1518248993, lastSeqNo 255958157, uniqueness 1406064021, timestamp 1407565307/15017590802014-08-09 14:21:48.052: CSSD3358clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958161, LATS 1518249002, lastSeqNo 255958158, uniqueness 1406064021, timestamp 1407565307/15017591912014-08-09 14:21:48.421: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 02014-08-09 14:21:49.043: CSSD4129clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958163, LATS 1518249993, lastSeqNo 255958160, uniqueness 1406064021, timestamp 1407565308/15017600822014-08-09 14:21:49.056: CSSD3358clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958164, LATS 1518250007, lastSeqNo 255958161, uniqueness 1406064021, timestamp 1407565308/15017601932014-08-09 14:21:49.421: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 02014-08-09 14:21:50.044: CSSD4129clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958166, LATS 1518250994, lastSeqNo 255958163, uniqueness 1406064021, timestamp 1407565309/15017610902014-08-09 14:21:50.057: CSSD3358clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958167, LATS 1518251007, lastSeqNo 255958164, uniqueness 1406064021, timestamp 1407565309/15017611952014-08-09 14:21:50.421: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 02014-08-09 14:21:51.046: CSSD4129clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958169, LATS 1518251996, lastSeqNo 255958166, uniqueness 1406064021, timestamp 1407565310/15017621002014-08-09 14:21:51.057: CSSD3358clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958170, LATS 1518252008, lastSeqNo 255958167, uniqueness 1406064021, timestamp 1407565310/15017622052014-08-09 14:21:51.102: CSSD5414clssnmSendingThread: sending join msg to all nodes2014-08-09 14:21:51.102: CSSD5414clssnmSendingThread: sent 5 join msgs to all nodes2014-08-09 14:21:51.421: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 02014-08-09 14:21:52.050: CSSD4129clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958172, LATS 1518253000, lastSeqNo 255958169, uniqueness 1406064021, timestamp 1407565311/15017631102014-08-09 14:21:52.058: CSSD3358clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958173, LATS 1518253008, lastSeqNo 255958170, uniqueness 1406064021, timestamp 1407565311/15017632302014-08-09 14:21:52.089: CSSD5671clssnmRcfgMgrThread: Local Join2014-08-09 14:21:52.089: CSSD5671clssnmLocalJoinEvent: begin on node(2), waittime 1930002014-08-09 14:21:52.089: CSSD5671clssnmLocalJoinEvent: set curtime (1518253039) for my node2014-08-09 14:21:52.089: CSSD5671clssnmLocalJoinEvent: scanning 32 nodes2014-08-09 14:21:52.089: CSSD5671clssnmLocalJoinEvent: Node rac01, number 1, is in an existing cluster with disk state 32014-08-09 14:21:52.090: CSSD5671clssnmLocalJoinEvent: takeover aborted due to cluster member node found on disk2014-08-09 14:21:52.431: CSSD4900clssgmWaitOnEventValue: after CmInfo State val 3, eval 1 waited 0从上面的信息，很容易给人感觉是心跳的问题。这么理解也不错，只是这里的心跳不是指的我们说理解的传统的心跳网络。我让他在crs正常的一个节点查询如下信息，我们就知道原因了，如下：SQLselectname,ip_addressfromv$cluster_interconnects;NAMEIP_ADDRESS-en0169.254.116.242大家可以看到，这里心跳IP为什么是169网段呢？很明显跟我们的/etc/hosts设置不匹配啊？why ？这里我们要介绍下Oracle 11gR2 引入的HAIP特性，Oracle引入该特性的目的是为了通过自身的技术来实现心跳网络的冗余，而不再依赖于第三方技术，比如Linux的bond等等。在Oracle 11.2.0.2版本之前，如果使用了OS级别的心跳网卡绑定，那么Oracle仍然以OS绑定的为准。从11.2.0.2开始，如果没有在OS层面进行心跳冗余的配置，那么Oracle自己的HAIP就启用了。所以你虽然设置的192.168.1.100，然而实际上Oracle使用是169.254这个网段。关于这一点，大家可以去看下alert log，从该日志都能看出来，这里不多说。我们可以看到，正常节点能看到如下的169网段的ip，问题节点确实看不到这个169的网段IP：Oracle MOS提供了一种解决方案，如下：crsctl start res ora.cluster_interconnect.haip -init经过测试，使用root进行操作，也是不行的。针对HAIP的无法启动，Oracle MOS文档说通常是如下几种情况：1） 心跳网卡异常2) 多播工作机制异常3）防火墙等原因4）Oracle bug对于心跳网卡异常，如果只有一块心跳网卡，那么ping其他的ip就可以进行验证了，这一点很容易排除。对于多播的问题，可以通过Oracle提供的mcasttest.pl脚本进行检测(请参考Grid Infrastructure Startup During Patching, Install or Upgrade May Fail Due to Multicasting Requirement (ID 1212703.1)，我这里的检查结果如下：$ ./mcasttest.pl -n rac02,rac01 -i en0# Setup for node rac02 #Checking node access rac02Checking node login rac02Checking/Creating Directory /tmp/mcasttest for binary on node rac02Distributing mcast2 binary to node rac02# Setup for node rac01 #Checking node access rac01Checking node login rac01Checking/Creating Directory /tmp/mcasttest for binary on node rac01Distributing mcast2 binary to node rac01# testing Multicast on all nodes #Test for Multicast address 230.0.1.0Aug 11 21:39:39 | Multicast Failed for en0 using address 230.0.1.0:42000Test for Multicast address 224.0.0.251Aug 11 21:40:09 | Multicast Failed for en0 using address 224.0.0.251:42001$虽然这里通过脚本检查，发现对于230和224网段都是不通的，然而这不见得一定说明是多播的问题导致的。虽然我们查看ocssd.log，通过搜索mcast关键可以看到相关的信息。实际上，我在自己的11.2.0.3 Linux RAC环境中测试，即使mcasttest.pl测试不通，也可以正常启动CRS的。由于网友这里是AIX，应该我就排除防火墙的问题了。因此最后怀疑Bug 9974223的可能性比较大。实际上，如果你去查询HAIP的相关信息，你会发现该特性其实存在不少的Oracle bug。其中 for knowns HAIP issues in 11gR2/12c Grid Infrastructure (1640865.1)就记录12个HAIP相关的bug。由于这里他的第1个节点无法操作，为了安全，是不能有太多的操作的。对于HAIP，如果没有使用多心跳网卡的情况下，我觉得完全是可以禁止掉的。但是昨天查MOS文档，具体说不能disabled。最后测试发现其实是可以禁止掉的。如下是我的测试过程：rootrac1 bin# ./crsctl modify res ora.cluster_interconnect.haip -attr ENABLED=0 -initrootrac1 bin# ./crsctl stop crsCRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on rac1CRS-2673: Attempting to stop ora.crsd on rac1CRS-2790: Starting shutdown of Cluster Ready Services-managed resources on rac1CRS-2673: Attempting to stop ora.oc4j on rac1CRS-2673: Attempting to stop ora.cvu on rac1CRS-2673: Attempting to stop ora.LISTENER_SCAN1.lsnr on rac1CRS-2673: Attempting to stop ora.GRID.dg on rac1CRS-2673: Attempting to stop ora.registry.acfs on rac1CRS-2673: Attempting to stop ora.rac1.vip on rac1CRS-2677: Stop of ora.rac1.vip on rac1 succeededCRS-2672: Attempting to start ora.rac1.vip on rac2CRS-2677: Stop of ora.LISTENER_SCAN1.lsnr on rac1 succeededCRS-2673: Attempting to stop ora.scan1.vip on rac1CRS-2677: Stop of ora.scan1.vip on rac1 succeededCRS-2672: Attempting to start ora.scan1.vip on rac2CRS-2676: Start of ora.rac1.vip on rac2 succeededCRS-2676: Start of ora.scan1.vip on rac2 succeededCRS-2672: Attempting to start ora.LISTENER_SCAN1.lsnr on rac2CRS-2676: Start of ora.LISTENER_SCAN1.lsnr on rac2 succeededCRS-2677: Stop of ora.registry.acfs on rac1 succeededCRS-2677: Stop of ora.oc4j on rac1 succeededCRS-2677: Stop of ora.cvu on rac1 succeededCRS-2677: Stop of ora.GRID.dg on rac1 succeededCRS-2673: Attempting to stop ora.asm on rac1CRS-2677: Stop of ora.asm on rac1 succeededCRS-2673: Attempting to stop ora.ons on rac1CRS-2677: Stop of ora.ons on rac1 succeededCRS-2673: Attempting to stop ora.net1.network on rac1CRS-2677: Stop of ora.net1.network on rac1 succeededCRS-2792: Shutdown of Cluster Ready Services-managed resources on rac1 has completedCRS-2677: Stop of ora.crsd on rac1 succeededCRS-2673: Attempting to stop ora.drivers.acfs on rac1CRS-2673: Attempting to stop ora.ctssd on rac1CRS-2673: Attempting to stop ora.evmd on rac1CRS-2673: Attempting to stop ora.asm on rac1CRS-2673: Attempting to stop ora.mdnsd on rac1CRS-2677: Stop of ora.mdnsd on rac1 succeededCRS-2677: Stop of ora.evmd on rac1 succeededCRS-2677: Stop of ora.ctssd on rac1 succeededCRS-2677: Stop of ora.asm on rac1 succeededCRS-2673: Attempting to stop ora.cluster_interconnect.haip on rac1CRS-2677: Stop of ora.cluster_interconnect.haip on rac1 succeededCRS-2673: Attempting to stop ora.cssd on rac1CRS-2677: Stop of ora.cssd on rac1 succeededCRS-2673: Attempting to stop ora.crf on rac1CRS-2677: Stop of ora.drivers.acfs on rac1 succeededCRS-2677: Stop of ora.crf on rac1 succeededCRS-2673: Attempting to stop ora.gipcd on rac1CRS-2677: Stop of ora.gipcd on rac1 succeededCRS-2673: Attempting to stop ora.gpnpd on rac1CRS-2677: Stop of ora.gpnpd on rac1 succeededCRS-2793: Shutdown of Oracle High Availability Services-managed resources on rac1 has completedCRS-4133: Oracle High Availability Services has been stopped.rootrac1 bin# ./crsctl start crsCRS-4123: Oracle High Availability Services has been started.rootrac1 bin# ./crsctl check crsCRS-4638: Oracle High Availability Services is onlineCRS-4537: Cluster Ready Services is onlineCRS-4529: Cluster Synchronization Services is onlineCRS-4533: Event Manager is onlinerootrac1 bin# ./crsctl stat res -t -init-NAME TARGET STATE SERVER STATE_DETAILS-Cluster Resources-ora.asm 1 ONLINE ONLINE rac1 Startedora.cluster_interconnect.haip 1 ONLINE OFFLINEora.crf 1 ONLINE ONLINE rac1ora.crsd 1 ONLINE ONLINE rac1ora.cssd 1 ONLINE ONLINE rac1ora.cssdmonitor 1 ONLINE ONLINE rac1ora.ctssd 1 ONLINE ONLINE rac1 ACTIVE:0ora.diskmon 1 OFFLINE OFFLINEora.drivers.acfs 1 ONLINE ONLINE rac1ora.evmd 1 ONLINE ONLINE rac1ora.gipcd 1 ONLINE ONLINE rac1ora.gpnpd 1 ONLINE ONLINE rac1ora.mdnsd 1 ONLINE ONLINE rac1rootrac1 bin#不过需要注意的是：当修改之后，两个节点都必须要重启CRS，否则仅仅重启一个节点的CRS是不行的，ASM实例是无法启动的。对于HAIP异常，为什么会导致节点的CRS无法正常启动呢？关于这一点，我们来看下该资源的属性就知道了，如下：NAME=ora.cluster_interconnect.haipTYPE=ora.haip.typeACL=owner:root:rw-,pgrp:oinstall:rw-,other:r-,user:oracle:r-xACTION_FAILURE_TEMPLATE=ACTION_SCRIPT=ACTIVE_PLACEMENT=0AGENT_FILENAME=%CRS_HOME%/bin/orarootagent%CRS_EXE_SUFFIX%AUTO_START=alwaysCARDINALITY=1CHECK_INTERVAL=30DEFAULT_TEMPLATE=DEGREE=1DESCRIPTION=Resource type for a Highly Available network IPENABLED=0FAILOVER_DELAY=0FAILURE_INTERVAL=0FAILURE_THRESHOLD=0HOSTING_MEMBERS=LOAD=1LOGGING_LEVEL=1NOT_RESTARTING_TEMPLATE=OFFLINE_CHECK_INTERVAL=0PLACEMENT=balancedPROFILE_CHANGE_TEMPLATE=RESTART_ATTEMPTS=5SCRIPT_TIMEOUT=60SERVER_POOLS=START_DEPENDENCIES=hard(ora.gpnpd,ora.cssd)pullup(ora.cssd)START_TIMEOUT=60STATE_CHANGE_TEMPLATE=STOP_DEPENDENCIES=hard(ora.cssd)STOP_TIMEOUT=0UPTIME_THRESHOLD=1mUSR_ORA_AUTO=USR_ORA_IF=USR_ORA_IF_GROUP=cluster_interconnectUSR_ORA_IF_THRESHOLD=20USR_ORA_NETMASK=USR_ORA_SUBNET=可以看到，如果该资源有异常，现在gpnpd,cssd都是有问题的。备注：实际上还可以通过在asm 层面指定cluster_interconnects来避免haip的问题。