2.10 What’s New (OES 2 SP3)

In addition to bug fixes and the changes from previously released patches, the following changes and enhancements were made for Novell Cluster Services for Linux in OES 2 SP3:

2.10.1 Support for the OES Common Proxy User

Support was added for the OES Common Proxy User feature in eDirectory 8.8.6. For information, see OES Common Proxy User.

2.10.2 NCS_Management Group

The NCS_Management group for a cluster contains the user name of the cluster administrator. This can be the OES Common Proxy user, an LDAP Admin user, or another administrator user that you specify for the cluster. Typically, the cluster administrator user is specified during the initial cluster configuration, as described in Step 3 of Section 5.5.5, Configuring a New Cluster. For information about configuring this user for an existing cluster, see Section 8.7, Changing the NCS Proxy User Assignment in the NCS_Management_Group.

2.10.3 Viewing and Configuring the LDAP Server Information

The following LDAP-related improvements were made for the handling of LDAP server information for a cluster:

  • Viewing the LDAP Server: A new option was added to show which LDAP server would be used if Novell Cluster Services initiated an LDAP transaction now. It also shows information about how Novell Cluster Services settled on the LDAP server (that is, the servers tried and results).

  • Preferred List for LDAP Servers: Novell Cluster Services switches to a different LDAP server if a dead server is encountered in the binding process. A list of LDAP servers that the cluster can use can be found in the /etc/opt/novell/ncs/clstrlib.conf file. Novell Cluster Services switches to the next LDAP server according to the order specified in the list.

2.10.4 Resource Mutual Exclusion Groups

The Resource Mutual Exclusion (RME) Groups feature allows you to configure sets of resources that cannot be assigned to the same node at the same time. For information, see Section 10.10, Configuring Resource Mutual Exclusion Groups.

2.10.5 Reboot Option for Failure Actions in Resource Monitoring

A new reboot option is available as a failure action in the Resource Monitoring setup. If a failure action initiates, this option reboots the operating system on the hosting node without synchronizing or unmounting the disks. Each of the resources on the hosting node fails over to the next server in its Assigned Nodes list because of the reboot. This is a hard reboot, not a graceful one. The reboot option is normally used only for a mission-critical cluster resource that must remain available. For information, see Failure Action.

2.10.6 Modifying the Preferred Nodes List as Text

A text editor environment is provided for ordering nodes listed in the Preferred Nodes page of the Clusters plug-in. Previously, you could modify the order of nodes in the list only by using the arrows to move one node up or down at a time. For information, see Step 7 in Section 10.8, Configuring Preferred Nodes for a Resource.

2.10.7 Modifying Resource Priorities List as Text

A text editor environment is provided for ordering resources in the Resource Priorities page of the Clusters plug-in. Previously, you could modify the order of resources in the list only by using the arrows to move one resource up or down at a time. For information, see Step 5 in Section 10.9, Configuring Resource Priorities for Load Order.

2.10.8 Cluster Report Information

In the Clusters plug-in for iManager, the Clusters > Cluster Manager > Run Report output has been modified to include the following information

  • Monitoring script for each cluster resource where Monitoring is enabled

  • An RME Groups section that lists the member resources of each group. For information, see Cluster Report.

2.10.9 Informative Subject Line for Email Notifications

The subject line of email messages sent by Novell Cluster Services now provides information about the cluster name, resource name, action taken, and node name. For example:

CL1: POOL1_SERVER online on NODE1

Previously, the subject line was Cluster Resource Change.

2.10.10 Confirming a Cluster Down or Cluster Restart Command

The cluster down command and cluster restart command now prompt for a confirmation (Yes/No) before bringing down the cluster.

2.10.11 Handling Reboot Preferences with the Linux Kernel Panic Setting

Novell Cluster Services has been modified to handle panic situations for a node according to the Linux kernel panic setting on the node. This setting determines whether the server is automatically rebooted after an automatic node shutdown. For information, see Section 9.11, Preventing a Cluster Node Reboot after a Node Shutdown.

2.10.12 Previewing All Cluster Resources Before the Final Conversion from NetWare 6.5 to OES 2 Linux

You can preview all resources before finalizing a cluster conversion by using the cluster convert preview command without specifying a resource name. Previously, you could preview only one resource at a time. For information, see Finalizing the Cluster Conversion in the OES 2 SP3: Novell Cluster Services NetWare to Linux Conversion Guide.

2.10.13 Providing the Password for ncs_install.py

The ncs_install.py script prompts for a password when it is run from the command line. Previously, the password was specified in the script.

2.10.14 Renaming a Pool Cluster Resource

The new cluster rename command allows you to rename a pool cluster resource.

cluster rename <resource_name> <new_resource_name>

For information, see Section 10.13, Renaming a Cluster Resource.

2.10.15 Controlling Monitoring for Resources

The new cluster monitor command allows you to start, stop, or view the status of monitoring for a specified cluster resource. Monitoring must be enabled for the cluster resource in order to use this command.

cluster monitor <resource_name> {start | stop | status}

For information, see Section 10.11, Controlling Resource Monitoring.

2.10.16 Cascading Failover Prevention

Novell Cluster Services added the cascading failover prevention function that detects if a node has failed because of a bad cluster resource and prevents that bad resource from failing over to other servers in the cluster. This function was previously available on NetWare, but not on Linux.