After upgrade to 5.1SP1, one node could not join the cluster. LLT startup failed with "cluster id already being used" error.

Problem

In a 2 node cluster, after upgrade to 5.1SP1, one node (nodeB) fails to join the cluster. LLT startup fails with error. The MAC address printed is the MAC address of one of the LLT links of its cluster partner (nodeA).

Error Message

[root:/etc/init.d]# ./llt.rc start
  Loading LLT Driver...
  Starting LLT...
  LLT lltconfig ERROR V-14-2-15245 cluster id 110 is already being used by nid 0 and has the address - XX:XX:XX:YY:YY:YY
  LLT lltconfig ERROR V-14-2-15664 LLT could not configure any link

Cause

In VCS 5.1SP1, duplicate cluster-id detection feature was introduced in LLT startup. The MAC address shown in the above error message is one of the private links of other node in cluster. To confirm this, the tcpdump was gathered for LLT links on both the nodes when LLT was started. In this case, 2 extra bytes were received for one of the LLT links which caused this bug to be hit.

Solution

It was suspected that the checksum on the NIC adapter puts the extra bytes on the packet. The LLT bug id for this issue is: 2439895 and is fixed in VCS 5.1SP1RP2. sort.symantec.com/patch/detail/5503


Applies To

Storage Foundation For  Oracle Real Application Cluster(SFROARAC) 5.1 SP1.

AIX 6.1

 

Terms of use for this information are found in Legal Notices.

Search

Survey

Did this article answer your question or resolve your issue?

No
Yes

Did this article save you the trouble of contacting technical support?

No
Yes

How can we make this article more helpful?

Email Address (Optional)