In this Document

Applies to:

Symptoms

During installation of 11.2.0.2 Grid Infrastructure software on SLES OS on IBM: Linux on System Z platform, the OUI runs the following command as a prereq check.

./runcluvfy
/oracle/oracle11g-software/grid/runcluvfy.sh stage -pre crsinst -n node1,node2 -verbose

The runcluvfy command is failing during nslookup as follows:-

Checking all nodes to make sure that search entry is "test.dom" as found on node "node1"
All nodes of the cluster have same value for 'search'

Checking DNS response time for an unreachable node
Node Name Status
------------------------------------ ------------------------
node1 failed
node2 passed
PRVF-5636 : The DNS response time for an unreachable node exceeded "15000" ms on following nodes: node1

File "/etc/resolv.conf" is not consistent across nodes

Check: Time zone consistency
Result: Time zone consistency check passed

Changes

10gR2 CRS currently installed.
11.2.0.2 Grid Infrastructure fresh installation.

Cause

There is an issue with the DNS configuration

runcluvfy is checking via nslookup that the response time for a "NON-EXISTENT" node returns an error within a specified time....

nslookup unknown e.g.

nslookup unknown
Server: 130.35.249.41
Address: 130.35.249.41#53

** server can't find unknown: NXDOMAIN


From the log above it shows that this is "successful" on node2 but fails on node1.

Further investigation shows the running
nslookup unknown on node1 just hangs and eventually returns connection timed out.

Solution

You need to work with your networking team and investigate the DNS setup.

On this occasion there was a problem with entries in the  /var/lib/named/root.hint of the dns Linux server.  Removing these entries resolved the "nslookup unknown" problem allowing runcluvfy to run successfully.