Sunday, July 17, 2011

root.sh failed at first node when installing 11gr2

Hello bussies,
                         Finally I got installed Oracle 11gR2 patchset 1 at AIX 6.1, the one which gave me hard time, and I was stuck in an issue for long time, I am sharing it with you guys.

                         During execution of root.sh at first node my configuration gets fail with below error

Start of resource "ora.cluster_interconnect.haip" failed
CRS-2672: Attempting to start 'ora.cluster_interconnect.haip' on 
'node1'
CRS-5017: The resource action "ora.cluster_interconnect.haip start" 
encountered the following error:
Start action for HAIP aborted
CRS-2674: Start of 'ora.cluster_interconnect.haip' on 'node1' failed
CRS-2679: Attempting to clean 'ora.cluster_interconnect.haip' on 'node1'
CRS-2681: Clean of 'ora.cluster_interconnect.haip' on 'mpdwsr01' succeeded
CRS-4000: Command Start failed, or completed with errors.
Failed to start Oracle Clusterware stack
Failed to start High Availability IP at 
/oracle/app/11.2.0/grid/crs/install/crsconfig_lib.pm line 1047.
/oracle/app/11.2.0/grid/perl/bin/perl -I/oracle/app/11.2.0/grid/perl/lib 
-I/oracle/app/11.2. 0/grid/crs/install 
/oracle/app/11.2.0/grid/crs/install/rootcrs.pl execution failed 


              Upon checking crsd.log I got an extract

2011-07-12 10:07:22.751: [GIPCXCPT][1543] gipchaInternalResolve: failed to resolve ret gipcretKeyNotFound (36), host 'ouptdbb01', port '14be-277c-83fd-56f3', hctx 110ea2210 [0000000000000010] { gipchaContext : host 'ouptdbb01', name 'a83b-2d41-7a3d-b0b6', luid '56bbeab0-00000000', numNode 0, numInf 1, usrFlags 0x0, flags 0x1 }, ret gipcretKeyNotFound (36)
             Orace support says its an undocumented bug 9593552 which is fixed in 11.2.0.2 PSU3, 11.2.0.3 and above, but unfortunately at time of this blog post I did not find its intermi patch 9593552 which Oracle says is the right solution.

            On further exploring this error I saw an abnormal behaviour of my private interconnect ethernet port that its going down during root.sh execution which iscausing the failure of starting ora.clusterinterconnect.haip,
I tried to start it manually using command 'crsctl start cluster' and good thing is that my ora.cluesterinterconnect.haip came up with out any issues and then I executed root.sh on second node and completed my 11gr2 patchset 1 grid installation.

            All srvices are up and running  on both nodes.

Regards.

Farrukh Salman,
Oracle DBA.

1 comment:

  1. Thanks! This helped me get my second node into the cluster. I had udp_sendspace set wrong too.

    ReplyDelete