listener process going down as VIP failed to start on primary node. (default gateway being changed or not found)

Post date: 22-May-2009 19:02:48

RAC uses default gateway as it's ping target, during VIP status / health check.

[oracle@rc1.myrac.com] # cst

HA Resource Target State

----------- ------ -----

ora.rc1.ASM1.asm ONLINE ONLINE on rc1

ora.rc1.LISTENER_DB_RC1.lsnr OFFLINE OFFLINE

ora.rc1.LISTENER_RC1.lsnr OFFLINE OFFLINE

ora.rc1.gsd ONLINE ONLINE on rc1

ora.rc1.ons ONLINE ONLINE on rc1

ora.rc1.vip ONLINE ONLINE on rc1

ora.rc2.ASM2.asm ONLINE ONLINE on rc2

ora.rc2.LISTENER_DB_RC2.lsnr ONLINE ONLINE on rc2

ora.rc2.LISTENER_RC2.lsnr ONLINE ONLINE on rc2

ora.rc2.gsd ONLINE ONLINE on rc2

ora.rc2.ons ONLINE ONLINE on rc2

ora.rc2.vip ONLINE ONLINE on rc2

ora.rdb.db ONLINE ONLINE on rc2

ora.rdb.rdb1.inst ONLINE ONLINE on rc1

ora.rdb.rdb2.inst ONLINE ONLINE on rc2

ora.rdb.serv1.cs ONLINE ONLINE on rc1

ora.rdb.serv1.rdb1.srv ONLINE ONLINE on rc1

[oracle@rc1.myrac.com] # srvctl start nodeapps -n rc1

rc1:ora.rc1.vip:checkIf: Default gateway is not defined (host=rc1.myrac.com)

rc1:ora.rc1.vip:Interface eth0 checked failed (host=rc1.myrac.com)

rc1:ora.rc1.vip:Invalid parameters, or failed to bring up VIP (host=rc1.myrac.com)

CRS-0215: Could not start resource 'ora.rc1.LISTENER_RC1.lsnr'.

As i already stated that server's default gateway is used as a ping target during the Oracle RAC 10g VIP status check action

The VIP as you can see for rc1 failed over to rc2 resulted in listener being down. This can be solved as follows :

1- stop nodeapps

2- As root,

vi the script $ORA_CRS_HOME/bin/racgvip and change the value of variable FAIL_WHEN_DEFAULTGW_NOT_FOUND=0 .

3- start nodeapps and you should see the resources ONLINE