I am having a very strange problem.
We have a 2 node cluster services with Netware 6.0 (I know 6.0 is no
longer support but I still need to fix my issue).
When I load LDNCS on the first server I get the message the node is a
member of the cluster but it doesn’t load/read the MasterIP resource. So
I have to manually load the commands in the MasterIP resource script
witch are the following:
set allow ip address duplicates = on
CLUSTER CVSBIND ADD ClusCP 172.17.45.180
NUDP ADD ClusCP 172.17.45.180
add secondary ipaddress 172.17.45.180
HTTPBIND 172.17.45.180 /KEYFILE:"SSL CertificateIP"
set allow ip address duplicates = off
As soon as the secondary ip address is added, cluster services seems to
kick in and reads the MasterIP script. However, because the ip address is
already added (manually by me) the script gets an error saying that the
ip address is already used. Because of this, the unload script is
automatically executed. So again, I manually add the secondary IP address
and then my cluster comes up but the MasterIP resource is in Comatose.
All other resources come up in normal condition. I then load LDNCS on the
other server and then I can migrate any service from one box to the other
without any problems.
However, if the node with the MasterIP resource crashes or abends, I have
to start the process all over.
Like I said, this is a weird one.