Building Disaster Recovery Serviceguard Solutions Using Metrocluster with Continuous Access EVA A.05.01

Check the package log files on all nodes of all other packages managed
by the Site Controller package to identify issues in those packages.
To clean a site for a site aware disaster tolerant application:
1. Clean the Site Safety Latch on the site by running the cmresetsc tool.
On a node from the site, run the following command:
# /usr/sbin/cmresetsc <Site_controller_package_name>
IMPORTANT: You must be a root user to run this command.
2. Check the package log file of the Site Controller Package on the node it failed on and fix any
reported issues.
3. Enable node switching for the Site Controller package on that node.
# cmmodpkg –e -n <node name> <site_controller_package_name>
4. Check the package log file on all nodes of the MNP packages managed by the Site Controller
package on the site.
5. Fix any issues reported in the package log files and enable node switching for the MNP
packages on the failed nodes.
# cmmodpkg -e -n <node 1 name> -n <node 2 name> <MNP Package name>
6. Restart the Site Controller package and enable global switching.
# cmrunpkg <site_controller_package_name>
# cmmodpkg -e <site_controller_package_name>
In addition to the cmresetsc tool, use the cmviewsc tool to view information on the Site Controller
package managed workload package. This tool is available in the /usr/sbin directory. Run the
following command to use the cmviewsc tool:
# /usr/sbin/cmviewsc [-v] [site_controller_package_name]
This cmviewsc tool displays the following information:
• Number of critical and managed packages at each site.
• Status of the Site Controller managed packages (halted or started).
• Site Controller managed packages halted cleanly or not.
• Site active or passive.
• Site Safety Latch value on each node. The value can be Close, Open, or Intermediate.
For more information on using cmviewsc, see cmviewsc(1m).
Identifying and cleaning MNP stack packages that are halted
The Site Controller package does not start if the MNP stack packages are not halted clean. An
MNP package is halted unclean when the halt script does not run successfully on all the configured
nodes of the package. This implies that there might be some stray resources configured with the
package, that are online in the cluster. The Site Controller package logs the following message in
its log file on the node where it failed to start:
Package <package name> has not halted cleanly on node <node name>
The following command shows whether an MNP package halt was clean or unclean:
# cmviewcl –v –f line <MNP Package name>
Check for the field last_halt_failed under each instance of the MNP package. When set to
Yes, that instance of the MNP package did not successfully run the halt script when it was halted.
Check for all instances.
70 Troubleshooting