HAIP异常，导致RAC节点无法启动的解决方案

http://ju.outofmemory.cn/entry/88036

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: [    CSSD][5414]clssnmSendingThread: sent 4 join msgs to all nodes
2014-08-09 14:21:46.421: [    CSSD][4900]clssgmWaitOnEventValue: after CmInfo State  val 3, eval 1 waited 0
2014-08-09 14:21:47.042: [    CSSD][4129]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958157, LATS 1518247992, lastSeqNo 255958154, uniqueness 1406064021, timestamp 1407565306/1501758072
2014-08-09 14:21:47.051: [    CSSD][3358]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958158, LATS 1518248002, lastSeqNo 255958155, uniqueness 1406064021, timestamp 1407565306/1501758190
2014-08-09 14:21:47.421: [    CSSD][4900]clssgmWaitOnEventValue: after CmInfo State  val 3, eval 1 waited 0
2014-08-09 14:21:48.042: [    CSSD][4129]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958160, LATS 1518248993, lastSeqNo 255958157, uniqueness 1406064021, timestamp 1407565307/1501759080
2014-08-09 14:21:48.052: [    CSSD][3358]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958161, LATS 1518249002, lastSeqNo 255958158, uniqueness 1406064021, timestamp 1407565307/1501759191
2014-08-09 14:21:48.421: [    CSSD][4900]clssgmWaitOnEventValue: after CmInfo State  val 3, eval 1 waited 0
2014-08-09 14:21:49.043: [    CSSD][4129]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958163, LATS 1518249993, lastSeqNo 255958160, uniqueness 1406064021, timestamp 1407565308/1501760082
2014-08-09 14:21:49.056: [    CSSD][3358]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958164, LATS 1518250007, lastSeqNo 255958161, uniqueness 1406064021, timestamp 1407565308/1501760193
2014-08-09 14:21:49.421: [    CSSD][4900]clssgmWaitOnEventValue: after CmInfo State  val 3, eval 1 waited 0
2014-08-09 14:21:50.044: [    CSSD][4129]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958166, LATS 1518250994, lastSeqNo 255958163, uniqueness 1406064021, timestamp 1407565309/1501761090
2014-08-09 14:21:50.057: [    CSSD][3358]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958167, LATS 1518251007, lastSeqNo 255958164, uniqueness 1406064021, timestamp 1407565309/1501761195
2014-08-09 14:21:50.421: [    CSSD][4900]clssgmWaitOnEventValue: after CmInfo State  val 3, eval 1 waited 0
2014-08-09 14:21:51.046: [    CSSD][4129]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958169, LATS 1518251996, lastSeqNo 255958166, uniqueness 1406064021, timestamp 1407565310/1501762100
2014-08-09 14:21:51.057: [    CSSD][3358]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958170, LATS 1518252008, lastSeqNo 255958167, uniqueness 1406064021, timestamp 1407565310/1501762205
2014-08-09 14:21:51.102: [    CSSD][5414]clssnmSendingThread: sending join msg to all nodes
2014-08-09 14:21:51.102: [    CSSD][5414]clssnmSendingThread: sent 5 join msgs to all nodes
2014-08-09 14:21:51.421: [    CSSD][4900]clssgmWaitOnEventValue: after CmInfo State  val 3, eval 1 waited 0
2014-08-09 14:21:52.050: [    CSSD][4129]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958172, LATS 1518253000, lastSeqNo 255958169, uniqueness 1406064021, timestamp 1407565311/1501763110
2014-08-09 14:21:52.058: [    CSSD][3358]clssnmvDHBValidateNCopy: node 1, rac01, has a disk HB, but no network HB, DHB has rcfg 217016033, wrtcnt, 255958173, LATS 1518253008, lastSeqNo 255958170, uniqueness 1406064021, timestamp 1407565311/1501763230
2014-08-09 14:21:52.089: [    CSSD][5671]clssnmRcfgMgrThread: Local Join
2014-08-09 14:21:52.089: [    CSSD][5671]clssnmLocalJoinEvent: begin on node(2), waittime 193000
2014-08-09 14:21:52.089: [    CSSD][5671]clssnmLocalJoinEvent: set curtime (1518253039) for my node
2014-08-09 14:21:52.089: [    CSSD][5671]clssnmLocalJoinEvent: scanning 32 nodes
2014-08-09 14:21:52.089: [    CSSD][5671]clssnmLocalJoinEvent: Node rac01, number 1, is in an existing cluster with disk state 3
2014-08-09 14:21:52.090: [    CSSD][5671]clssnmLocalJoinEvent: takeover aborted due to cluster member node found on disk
2014-08-09 14:21:52.431: [    CSSD][4900]clssgmWaitOnEventValue: 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：

SQL> select name,ip_address from v$cluster_interconnects;
NAME            IP_ADDRESS
--------------- ----------------
en0             169.254.116.242

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 'rac02'
Checking node login 'rac02'
Checking/Creating Directory /tmp/mcasttest for binary on node 'rac02'
Distributing mcast2 binary to node 'rac02'
###########  Setup for node rac01  ##########
Checking node access 'rac01'
Checking node login 'rac01'
Checking/Creating Directory /tmp/mcasttest for binary on node 'rac01'
Distributing mcast2 binary to node 'rac01'
###########  testing Multicast on all nodes  ##########
Test for Multicast address 230.0.1.0
Aug 11 21:39:39 | Multicast Failed for en0 using address 230.0.1.0:42000
Test for Multicast address 224.0.0.251
Aug 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：

[root@rac1 bin]# ./crsctl modify res ora.cluster_interconnect.haip -attr "ENABLED=0" -init
[root@rac1 bin]# ./crsctl stop crs
CRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on 'rac1'
CRS-2673: Attempting to stop 'ora.crsd' on 'rac1'
CRS-2790: Starting shutdown of Cluster Ready Services-managed resources on 'rac1'
CRS-2673: Attempting to stop 'ora.oc4j' on 'rac1'
CRS-2673: Attempting to stop 'ora.cvu' on 'rac1'
CRS-2673: Attempting to stop 'ora.LISTENER_SCAN1.lsnr' on 'rac1'
CRS-2673: Attempting to stop 'ora.GRID.dg' on 'rac1'
CRS-2673: Attempting to stop 'ora.registry.acfs' on 'rac1'
CRS-2673: Attempting to stop 'ora.rac1.vip' on 'rac1'
CRS-2677: Stop of 'ora.rac1.vip' on 'rac1' succeeded
CRS-2672: Attempting to start 'ora.rac1.vip' on 'rac2'
CRS-2677: Stop of 'ora.LISTENER_SCAN1.lsnr' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.scan1.vip' on 'rac1'
CRS-2677: Stop of 'ora.scan1.vip' on 'rac1' succeeded
CRS-2672: Attempting to start 'ora.scan1.vip' on 'rac2'
CRS-2676: Start of 'ora.rac1.vip' on 'rac2' succeeded
CRS-2676: Start of 'ora.scan1.vip' on 'rac2' succeeded
CRS-2672: Attempting to start 'ora.LISTENER_SCAN1.lsnr' on 'rac2'
CRS-2676: Start of 'ora.LISTENER_SCAN1.lsnr' on 'rac2' succeeded
CRS-2677: Stop of 'ora.registry.acfs' on 'rac1' succeeded
CRS-2677: Stop of 'ora.oc4j' on 'rac1' succeeded
CRS-2677: Stop of 'ora.cvu' on 'rac1' succeeded
CRS-2677: Stop of 'ora.GRID.dg' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.asm' on 'rac1'
CRS-2677: Stop of 'ora.asm' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.ons' on 'rac1'
CRS-2677: Stop of 'ora.ons' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.net1.network' on 'rac1'
CRS-2677: Stop of 'ora.net1.network' on 'rac1' succeeded
CRS-2792: Shutdown of Cluster Ready Services-managed resources on 'rac1' has completed
CRS-2677: Stop of 'ora.crsd' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.drivers.acfs' on 'rac1'
CRS-2673: Attempting to stop 'ora.ctssd' on 'rac1'
CRS-2673: Attempting to stop 'ora.evmd' on 'rac1'
CRS-2673: Attempting to stop 'ora.asm' on 'rac1'
CRS-2673: Attempting to stop 'ora.mdnsd' on 'rac1'
CRS-2677: Stop of 'ora.mdnsd' on 'rac1' succeeded
CRS-2677: Stop of 'ora.evmd' on 'rac1' succeeded
CRS-2677: Stop of 'ora.ctssd' on 'rac1' succeeded
CRS-2677: Stop of 'ora.asm' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.cluster_interconnect.haip' on 'rac1'
CRS-2677: Stop of 'ora.cluster_interconnect.haip' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.cssd' on 'rac1'
CRS-2677: Stop of 'ora.cssd' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.crf' on 'rac1'
CRS-2677: Stop of 'ora.drivers.acfs' on 'rac1' succeeded
CRS-2677: Stop of 'ora.crf' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.gipcd' on 'rac1'
CRS-2677: Stop of 'ora.gipcd' on 'rac1' succeeded
CRS-2673: Attempting to stop 'ora.gpnpd' on 'rac1'
CRS-2677: Stop of 'ora.gpnpd' on 'rac1' succeeded
CRS-2793: Shutdown of Oracle High Availability Services-managed resources on 'rac1' has completed
CRS-4133: Oracle High Availability Services has been stopped.
[root@rac1 bin]# ./crsctl start crs
CRS-4123: Oracle High Availability Services has been started.
[root@rac1 bin]# ./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
[root@rac1 bin]# ./crsctl stat res -t -init
--------------------------------------------------------------------------------
NAME           TARGET  STATE        SERVER                   STATE_DETAILS
--------------------------------------------------------------------------------
Cluster Resources
--------------------------------------------------------------------------------
ora.asm
      1        ONLINE  ONLINE       rac1                     Started
ora.cluster_interconnect.haip
      1        ONLINE  OFFLINE
ora.crf
      1        ONLINE  ONLINE       rac1
ora.crsd
      1        ONLINE  ONLINE       rac1
ora.cssd
      1        ONLINE  ONLINE       rac1
ora.cssdmonitor
      1        ONLINE  ONLINE       rac1
ora.ctssd
      1        ONLINE  ONLINE       rac1                     ACTIVE:0
ora.diskmon
      1        OFFLINE OFFLINE
ora.drivers.acfs
      1        ONLINE  ONLINE       rac1
ora.evmd
      1        ONLINE  ONLINE       rac1
ora.gipcd
      1        ONLINE  ONLINE       rac1
ora.gpnpd
      1        ONLINE  ONLINE       rac1
ora.mdnsd
      1        ONLINE  ONLINE       rac1
[root@rac1 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.haip
TYPE=ora.haip.type
ACL=owner:root:rw-,pgrp:oinstall:rw-,other::r--,user:oracle:r-x
ACTION_FAILURE_TEMPLATE=
ACTION_SCRIPT=
ACTIVE_PLACEMENT=0
AGENT_FILENAME=%CRS_HOME%/bin/orarootagent%CRS_EXE_SUFFIX%
AUTO_START=always
CARDINALITY=1
CHECK_INTERVAL=30
DEFAULT_TEMPLATE=
DEGREE=1
DESCRIPTION="Resource type for a Highly Available network IP"
ENABLED=0
FAILOVER_DELAY=0
FAILURE_INTERVAL=0
FAILURE_THRESHOLD=0
HOSTING_MEMBERS=
LOAD=1
LOGGING_LEVEL=1
NOT_RESTARTING_TEMPLATE=
OFFLINE_CHECK_INTERVAL=0
PLACEMENT=balanced
PROFILE_CHANGE_TEMPLATE=
RESTART_ATTEMPTS=5
SCRIPT_TIMEOUT=60
SERVER_POOLS=
START_DEPENDENCIES=hard(ora.gpnpd,ora.cssd)pullup(ora.cssd)
START_TIMEOUT=60
STATE_CHANGE_TEMPLATE=
STOP_DEPENDENCIES=hard(ora.cssd)
STOP_TIMEOUT=0
UPTIME_THRESHOLD=1m
USR_ORA_AUTO=
USR_ORA_IF=
USR_ORA_IF_GROUP=cluster_interconnect
USR_ORA_IF_THRESHOLD=20
USR_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。

Related posts:

1. 关于oracle rac节点的驱逐
2. Linux rac delete node小记
3. oracle restart体验（续）
4. oracle 诊断利器之一—Procwatcher
5. oracle restart–apply psu(11.2.0.2.4)