Node is refusing Serviceguard Communication

[Lee Harris_5]

Hi,

I've read a number of other threads on the forums about this problem, but none of the usual things seem to have helped me. I'm trying to create a new two node cluster. Both boxes are using /etc/hosts for name resolution, both can nslookup and ping each other and have the relevant entries in /etc/hosts. The cmclnodelist file contains the host names, and the cmclconfig.ascii contains the correct hostnames. I can rcp and remsh between the boxes in either direction as root. .rhosts allows root unrestricted access between the two nodes, but each time i try to run cmcheckconf -v -C cmclconfig.ascii from either node I get the message...

Checking cluster file: cmclconfig.ascii
Checking nodes ... Done
Checking existing configuration ... Done
Warning: Unable to get configuration for cluster dsg_test_cluster1.
Error: Node myserver1 is refusing Serviceguard communication.
Please make sure that the proper security access is configured on node
myserver1 through either file-based access (pre-A.11.16 version) or role-based
access (version A.11.16 or higher) and/or that the host name lookup
on node myserver1 resolves the IP address correctly.
cmcheckconf : Failed to gather configuration information

I'm using A.11.16 of SG on HP-UX 11.11. I have also just recently installed the patch PHSS_32731 and PHSS_32732.

Any ideas?

Many thanks

[Ninad_1]

Lee,

It seems you have not configured the access rights - either by
1. Create a /etc/cmcluster/cmclnodelist
and add the following entries
node1 root
node2 root

Assuming you have 2 nodes - node1 and node2 in your cluster

OR

2. In file /etc/hosts.equiv add
node1 root
node2 root

Ans then try.
Also read the Managing Service Guard Manual

Regards,
Ninad

[Ninad_1]

One more thing, also check your /etc/nsswitch.conf , /etc/hosts files to ensure you have proper host name resolution service running and that the nodes are accessible using there node names.
Also find the link for the manual
http://docs.hp.com/en/B3936-90079/index.html

Regards,
Ninad

[Luk Vandenbussche]

Hi,

Check the release notes to see if all the required patches are installed

http://docs.hp.com/en/B3935-90078/B3935-90078.pdf

[Lee Harris_5]

Hi,

I have an /etc/cmcluster/cmclnodelist on both servers with the hostnames as you mentioed. I've created /etc/hosts.equiv on both servers. I've check /etc/resolv.conf and /etc/nsswitch.conf and all of this is fine. I also added an access policy to cmclconfig.ascii of

USER_NAME root
USER_HOST ANY_SERVICEGUARD_NODE
USER_ROLE FULL_ADMIN

Now when I run cmcheckconf -v -C cmclconfig.ascii i get the following...

Checking cluster file: cmclconfig.ascii
Checking nodes ... Done
Checking existing configuration ... Done
Warning: Unable to get configuration for cluster sgtest1.
Error: Node myserver1 is refusing Serviceguard communication.
Please make sure that the proper security access is configured on node
myserver1 through either file-based access (pre-A.11.16 version) or role-based
access (version A.11.16 or higher) and/or that the host name lookup
on node myserver1 resolves the IP address correctly.
Error: Unable to get cluster handle for cluster sgtest1in cf_read_cluster_file.
cmcheckconf : Error found in cluster file: cmclconfig.ascii.

[LiPEnS]

Hi
check this http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=977854

Regards LiPEnS

[Lee Harris_5]

Hi all,

Eventually got this working. In A.11.16 cmclnodelist is obsolete. Policies for access control are defined in the cluster configuration. However before the cluster is built, there is obviously no configuration therefore no access policies are setup. As such, initially cmclnodelist is used as a bootstrap to build the first configuration. All I did was change cmclnodelist on both nodes to contain + and that seemed to do the trick.

Thanks for your help.