This section explains advisory notes regarding the settings for managers in cluster operations, and their deletion.
Switching Cluster Services (Cluster Applications)
Events that occur when switching cluster services (cluster applications) cannot be displayed.
Also, the message number 65529 may be displayed on the ROR console.
Perform login again.
For details on the messages, refer to "Message number 65529" in "Messages".
Troubleshooting Information
Collect troubleshooting information referring to the instructions in "1.2 Collecting Troubleshooting Data (Virtual Edition)" in "Troubleshooting". At that time, execute the commands from the primary node of the admin server, and collect information from the primary node and managed servers.
Commands
Do not use the start or stop subcommand of rcxadm mgrctl to start or stop the manager.
Right-click the manager "service or application" on the Failover Cluster Management tree, and select [Bring this service or application online] or [Take this service or application offline] from the displayed menu.
Other commands can be executed as they are in normal cluster operation.
ROR Console
The registration state of server resources on the resource tree when an admin server is included in a chassis is as indicated below.
The server resources being operated by the manager on the cluster node are displayed as "[Admin Server]".
The server resources not being operated by the manager on the cluster node are displayed as "[Unregistered]".
Do not register the server resources of the cluster node that are displayed as "[Unregistered]".
Services (Daemons) Managed with PRIMECLUSTER [Linux Manager]
The following services (daemons) are managed with the Check script provided by Resource Orchestrator.
/etc/init.d/scwdepsvd
/etc/init.d/scwpxesvd
/etc/init.d/scwtftpd
/opt/FJSVrcvmr/opt/FJSVssmgr/bin/cimserver
/opt/FJSVrcvmr/sys/rcxtaskmgr
/opt/FJSVrcvmr/sys/rcxmongrel1
/opt/FJSVrcvmr/sys/rcxmongrel2
/opt/FJSVrcvmr/sys/rcxhttpd
/etc/init.d/dhcpd (*)
* Note: When registering the admin LAN subnet
