The warning messages of the userApplication Configuration Wizard GUI are described in message number sequence.
Content:
Since other client is using userApplication Configuration Wizard GUI or the hvw(1M), GUI cannot be started.
Corrective action:
Check whether the userApplication Configuration Wizard GUI has already been started in another client or whether another client is using the hvw command. After the other operation is completed, start the GUI or command again.
This message may be displayed at the next startup even if the Web browser or the cluster management server of Web-Based Admin View is restarted while the GUI is being displayed. If this is the case, wait about five minutes, and then restart the Web browser and the GUI.
Content:
Cluster resource management facility is not running. Since a list of candidate interfaces cannot be obtained, GUI is terminated.
Corrective action:
The reasons for this event occurring includes the following:
Cause 1
A node for which initial setup of the cluster resource management facility has not yet been completed is included among the nodes to be monitored with Web-Based Admin View.
Cause 2
A node on which the cluster resource management facility is not running is included among the nodes to be monitored with Web-Based Admin View.
Cause 3
The software that uses the network service may not be able to start due to an error in the use /etc/inet/services file.
How to confirm cause 1
Confirm whether a node for which initial setup of the cluster resource management facility has not yet been completed is included among the nodes to be monitored with Web-Based Admin View.
Make the confirmation as follows:
Confirm the nodes to be monitored with Web-Based Admin View.
Execute the wvstat (1M) command to confirm the monitoring-target nodes for each operation management server.
Confirm the settings of the resource database.
Use the CRM main window of Cluster Admin or the clgettree (1) command to confirm whether all the nodes that were confirmed in Step 1 can be displayed.
If Web-Based Admin View uses a network other than a public LAN for operation management, the node names displayed in Steps 1 and 2 will be different.
Corrective action for cause 1
If only initial setup of the cluster resource management facility has not been completed, use the userApplication Configuration Wizard GUI after performing initial setup and automatic configuration on the node on which initial setup of the cluster resource management facility has not been completed.
For details on initial setup of the cluster resource management facility, see "PRIMECLUSTER Installation and Administration Guide."
If installation of the product package has been completed, use the userApplication Configuration Wizard GUI after node expansion, as explained in "Adding a Node."
In all other cases, for the node on which initial setup of the cluster resource management facility has not been completed, change the settings of the operation management server of Web-Based Admin View to a host that is not the current operation management server.
For details on how to change the operation management server of Web-Based Admin View, see "Initial setup of the operation management server" in "PRIMECLUSTER Installation and Administration Guide."
How to confirm cause 2
Confirm whether the node on which the cluster resource management facility is not operating is contained among the Web-Based Admin View monitoring-target nodes.
Make the confirmation as follows:
Confirm the nodes to be monitored with Web-Based Admin View.
Execute the wvstat (1M) command to confirm the monitoring-target nodes for each operation management server.
Confirm the operation status of the cluster resource management facility.
Use the CRM main window of Cluster Admin or the clgettree (1) command to confirm which node is in a state other than "ON."
Corrective action for cause 2
Boot the node on which the cluster resource management facility is not operating.
Corrective action for cause 3
Correct the /etc/inet/services file.
Content:
RMS is running. Generation and distribution of the Configuration information may fail. Thus it may not be created or changed.
Corrective action:
Stop RMS, and then re-execute the operation.
Content:
An invalid character is included in the cluster application name, resource name, script name, and script path.
Corrective action:
A cluster application name, a resource name, a script name, or a script path includes an inappropriate character. Re-enter the correct value.
Content:
All resources under the cluster application are to be removed.
Corrective action:
To remove the cluster application indicated by "userApplication name" in the message and all resources (resource name) under that cluster application, select Yes. If you do not want to execute deletion processing, select No.
Content:
The cluster application, resource name, or script name has not been entered.
Corrective action:
Enter the cluster application name, resource name, or script name.
Content:
The numbers that were entered for TIMEOUT, netmask, prefix-length, and so on are invalid.
Corrective action:
Re-enter an appropriate value.
Content:
The specified takeover IP address is not available.
Corrective action:
Enter another takeover IP address.
Content:
There is an incorrect content in set item.
Corrective action:
Change the attribute (that has the incorrect data), and continue the operation.
Content:
The takeover network information has already been described in the /usr/opt/reliant/etc/hvipalias file.
Corrective action:
To use the values that were set, select Yes.
If you select No, the userApplication Configuration Wizard GUI re-creates the takeover network information in the previous file and in /etc/inet/hosts. In this case, the takeover network information can be deleted automatically from /usr/opt/reliant/etc/hvipalias and /etc/inet/hosts when the Ipaddress resource is deleted.
Content:
There is an attribute different than the ones of other resources. This message asks user to continue the operation.
Corrective action:
Review the displayed attribute for any problems. If you select Yes, the system sets the attribute information as specified.
Content:
The file name is not specified.
Corrective action:
Enter the path of the reference file.
Content:
A required setup is missing.
Corrective action:
Set the required items in the screen.
Content:
The combination of cluster applications that the exclusion relationship was set is not correct.
Corrective action:
You need to set two or more cluster applications in one exclusion group.
Content:
The selected userApplication or Resource cannot be edited.
Corrective action:
Change the cluster application name or the Resource name, and execute the setup process again.
Content:
The specified takeover IP address or host name has been used.
Corrective action:
Enter another IP address or host name.
Content:
Invalid file name or path.
Corrective action:
Enter another reference file path.
Content:
The specified file exists.
Corrective action:
If you select Yes, the existing file will be replaced. If you select No, the existing file will not be replaced.
Content:
The IP address being allocated to the network interface card to be used in the takeover network is different.
Corrective action:
To continue the setup with the assigned IP address segment, select Yes. To select another NIC, select No.
Content:
The file system has been used.
Corrective action:
Set another device path and mount point.
Content:
The candidate interfaces to be set to resources cannot be obtained.
Corrective action:
Check whether the interfaces were previously registered to the cluster resource management facility, and then re-execute the operation.
Content:
The specified file cannot be executed.
Corrective action:
Use another file, or add an execute permission to the file.
Content:
The configuration is not set to use I/O fencing.
Corrective action:
When using I/O fencing, check the [I/O fencing] check box of [Attributes] tab.
When not using I/O fencing, make sure to refer to "Supplement on I/O fencing resource deletion" in "PRIMECLUSTER Installation and Administration Guide (Oracle Solaris)" and review the configuration and the settings.
Content:
The setting of GDS/SF to be used for I/O fencing cannot be executed.
Corrective action:
When using I/O fencing, review the following settings:
[SDX_VM_IO_FENCE=on] is set in all the nodes of /etc/opt/FJSVsdx/sdx.cf file
Settings of XSCF (SPARC M10/M12) or ICMP shutdown agent
For details on the settings of [SDX_VM_IO_FENCE=on], see "Setting I/O Fencing Function of GDS" in "PRIMECLUSTER Installation and Administration Guide (Oracle Solaris)."
For details on the settings of XSCF (SPARC M10/M12) and ICMP shutdown agent, see "Configuring the Shutdown Facility" in "PRIMECLUSTER Installation and Administration Guide (Oracle Solaris)."
Content:
Unable to operate multiple userApplication in the environment which uses ICMP shutdown agent.
Corrective action:
If you can change the shutdown agent to use XSCF (SPARC M10/M12), select Yes, and then proceed with the creating of userApplication.
If you cannot change the shutdown agent to use XSCF (SPARC M10/M12), select No, and then stop with the creating of userApplication.
Register also all of the resources to one of the userApplication.
For details on the settings of XSCF (SPARC M10/M12) shutdown agent, see "Configuring the Shutdown Facility" in "PRIMECLUSTER Installation and Administration Guide (Oracle Solaris)."
