Chapter Review Questions

     
1:

I am currently using the default failover and failback policies for my applications. I want to utilize a Rolling Standby configuration. What configuration changes do I need to make to my cluster and application configuration files? Will these changes require the cluster and/or the package to be halted? Are there any implications for the overall configuration of the individual nodes in the cluster by implementing a Rolling Standby configuration?

2:

Your Primary LAN card fails. An engineer has replaced the card. The card has maintained its Instance number and associated device files. You are not going to change the IP address assigned to this LAN card. Will this node be able to rejoin the cluster with no further work except to use the cmrunnode command? If not, what changes to the cluster and application configuration files will you need to make? Are there any other commands you need to run or changes you must make before allowing this node to rejoin the cluster?

3:

You have found on one node in your cluster an ASCII file that describes the cluster configuration. You are about to add a new node to the cluster. Is it safe to use the ASCII file you have found? If not, what should you do in order to acquire the most up-to-date cluster configuration in ASCII file format?

4:

I have removed a package from the cluster using the following checklist:

  1. Halt the package.

  2. Remove the entire package directory ( /etc/cmcluster/mypackage ) from the /etc/cmluster directory.

  3. Recompile the cluster binary including all existing packages and leaving out the package I have just removed.

Is this checklist sufficient in order to completely remove the package from the cluster configuration? If not, what is wrong with the checklist and what should the checklist look like?

5:

On my four-node cluster, one node has failed and the machine itself is halted. The node has experienced a series of catastrophic hardware failures and is currently going through a series of rigorous diagnostic checks by HP hardware engineers . The initial diagnosis is that it may be days or even weeks before all the relevant parts are shipped to site, and even then it may take some time to re-commission the machine because further diagnostic stress tests are required to ensure that the machine is functioning as normal. We decide to remove the node from the cluster. Can I do this while the cluster is down? Explain your answer. Are there any subsequent configuration changes that we may want to consider as a result of removing this node from the cluster? Explain your answer.



HP-UX CSE(c) Official Study Guide and Desk Reference
HP-UX CSE(c) Official Study Guide and Desk Reference
ISBN: N/A
EAN: N/A
Year: 2006
Pages: 434

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net