Sun Cluster Geographic Edition 3.2 Release Notes

Known Issues and Bugs

The following known issues and bugs affect the operation of the Sun Cluster Geographic Edition 3.2 release.

No RBAC Support for Sun Cluster Manager (6226493)

Problem Summary: Sun Cluster Manager does not support RBAC roles.

Workaround: Invoke Sun Cluster Manager as superuser on the local cluster.

Sun Cluster Manager Requires Same Root Password on Partner Clusters (6260505)

Problem Summary: To use the Sun Cluster Manager graphical user interface (GUI), the root password must be the same on all nodes of both clusters in the Sun Cluster Geographic Edition deployment.

Workaround: If you use Sun Cluster Manager to configure your clusters, ensure that the root password is the same on every node of both clusters. If you prefer to not set the root password identically on all nodes, use the command-line interface to configure your clusters.

A Custom Heartbeat Must Exist on Both the Remote and Local Cluster Before the Heartbeat Can Join a Partnership (6263692)

Problem Summary: If a partnership is created on a remote cluster by using a custom heartbeat, then a heartbeat by the same name must exist on the local cluster before it can join the partnership. You cannot create a heartbeat by using the GUI, so the appropriate heartbeat will not be available to choose on the Join Partnership page.

Workaround: Use the command-line interface (CLI) to create the custom heartbeat, and then use either the CLI or Sun Cluster Manager to join the partnership.

Sun Cluster Manager Does Not Always Refresh Protection Group Status Change (6302217)

Problem Summary: Configuration and state changes of entities on a page that is displayed in Sun Cluster Manager should cause the page to be refreshed automatically. Sometimes the refresh does not take place.

Workaround: Use the navigation tree to navigate to a different page, then return to the original page. It will be refreshed on reload.

`tcp_udp_plugin` Reports `No_Response` on Cluster Start (6412025)

Problem Summary: When restarting the clusters, the heartbeat is in the Degraded state and the Plug-in tcp_udp_plugin is in the No_Response state. On the partner cluster, the process tcp_udp_resp does not exist.

Workaround: Restart the tcp_udp_resp process on the partner cluster by issuing pkill -9 tcp_udp_resp on the partner cluster.

Neither `geoadm start` Nor `geoadm stop` Function Properly When Sun Cluster Geographic Edition Resources are in the `STOP_FAILED` State (6446765)

Problem Summary: It is possible that certain unusual configuration errors could leave the cluster in a state where the Sun Cluster Geographic Edition framework can neither be started (geoadm start) nor cleanly stopped (geoadm stop).

Workaround: It is most likely that a Sun Cluster Geographic Edition infrastructure resource is in the STOP_FAILED state. To clear the STOP_FAILED state, take the following actions:

Use the scstat -g command to determine which resources and resource groups are affected.
Clear the STOP_FAILED flag for all resources and resource groups that are in the STOP_FAILED state by using the following command for each:
# scswitch -c -j resource -h nodename -f STOP_FAILED
Manually stop the application that failed to stop.

For example, if an ora lsnr failed to stop, then stop it fully. Ignore this step if the affected resources are Sun Cluster Geographic Edition infrastructure only.
If necessary, stop the resource groups.

If a resource failed to stop during a resource group stop, then the resource group remains in the STOP_FAILED state and you must stop it by using the following command:
# scswitch -F -g resourcegroup
If the resources failed to stop during a restart of the resource or while the resource was being disabled then ignore this step.
Retry the geoadm stop command.

When Using EMC Symmetrix for Data Replication, `geopg switchover` Command Fails (6456435)

Problem Summary: EMC SRDF does not allow role changes when domino effect is on. If adaptive copy write pending and domino modes are set, you see failure messages that are similar to the following example:

# geopg switchover -f -m no-1 srdfpg
Processing operation... The timeout period for this operation on each cluster is 3600 seconds 
(3600000 milliseconds)...

Resource groups "fo-rg,scal-rg" is unmanaged.
Failed to perform the protection group role change for device group  testdg.

# Aug  2 11:05:14 pbrew2 symrdf[29990]: 'Suspend' for device group testdg - Cannot proceed in the 
current RDF state except if the Symmetrix force flag is used. Note that the Symmetrix force flag is 
not allowed to be used by default in RDF control operations

Aug  2 11:05:17 pbrew2 symrdf[11]: 'Swap Personality' for device group testdg - The device is not 
in a valid RDF state for this operation

Workaround: Use failover instead of switchover.

`gchb_resd` Dumps Core During `geoadm start` After Software is Removed and Reinstalled (6473292)

Problem Summary: If you use pkgrm to remove the Sun Cluster Geographic Edition software, the gchb_resd process might be left running. In this case, if you then reinstall, the process crashes.

Workaround: None needed. The gchb_resd process restarts automatically.

Unable to Install on a Cluster That is Upgraded to Sun Cluster 3.2 (6474269)

Problem Summary: On a cluster that is upgraded to Sun Cluster 3.2 software, the installer is unable to install Sun Cluster Geographic Edition 3.2 software. Because Sun Cluster software is not upgraded with the installer program, the installer program cannot recognize that the required Sun Cluster 3.2 software is running on the cluster.

Workaround: Perform the following steps on each node.

Rename the product registry file.

# mv /var/sadm/install/productregistry /var/sadm/install/productregistry_$date

Remove the Java^TM Enterprise System installation and uninstallation packages.
# pkgrm SUNWentsys5 SUNWentsys5i SUNWentsys4 SUNWentsys4i
Rerun the installer program to install Sun Cluster Geographic Edition 3.2 software.

Uninstalling and Reinstalling Fails (6489051)

Problem Summary: If you uninstall Sun Cluster Geographic Edition then reinstall it, the enabling of Sun Cluster Geographic Edition software fails. Messages are similar to the following example:

# geoadm start
... checking for management agent ...
... management agent check done ....
... starting product infrastructure ... please wait ...
Nov  1 14:35:16 phys-obsidian-1 Cluster.GCHB_resd: GCHB system error: pmfadm failed - 
could not start agent /usr/cluster/lib/geo/lib/tcp_udp_resp
Nov  1 14:36:54 phys-obsidian-1 SC[SUNW.scmasa,geo-infrastructure,geo-failovercontrol,scmasa_svc_start]: 
Failed to start /usr/cluster/lib/rgm/rt/hamasa/cmas_service_ctrl_start geo-infrastructure.
Nov  1 14:36:54 phys-obsidian-1 Cluster.RGM.rgmd: Method <scmasa_svc_start> failed on resource 
<geo-failovercontrol> in resource group <geo-infrastructure> [exit code <50>, 
time used: 21% of timeout <600 seconds>] 
Nov  1 14:37:22 phys-obsidian-1 Cluster.PMF.pmfd: chdir: No such file or directory

Workaround: After you reinstall Sun Cluster Geographic Edition software, restart the common agent container.

# /usr/sbin/cacaoadm restart