Showing results for 
Search instead for 
Do you mean 

Trucluster connection manager

Occasional Visitor

Trucluster connection manager

We have a 2 member cluster (5.1a), however when we simulate a power loss ( switch power off on one of the members), the remaining cluster member freezes. We see the message "CNX MGR: delay 121 secs 0 usecs". I cannot find anything that describes how this 121 second delay is calculated.

3 REPLIES
Respected Contributor

Re: Trucluster connection manager

Peter,

You can tune the cluster_transition_time.
In following document, you'll find all about.
http://h30097.www3.hp.com/docs/best_practices/BP_CLU/bp_clu.pdf

Rgrds

Joris
To err is human, but to really faul things up requires a computer
Honored Contributor

Re: Trucluster connection manager

The message is normaly bundled with the following additional messages:

CNX MGR: delay 1 secs 0 usecs
CNX QDISK: Cluster transition, releasing claim to 1 quorum disk vote.
CNX MGR: quorum lost, suspending cluster operations.

so it seems your machine is currently in state of cluster transition failing over to the survived member if quorum is correctly configured.
Help() { FirstReadManual(urgently); Go_to_it;; }
Occasional Visitor

Re: Trucluster connection manager

Joris,

Thanks for that, it was very usefull.

Pete
//Add this to "OnDomLoad" event