Novell Cluster Services 2.0 supports OES 11 services and file systems running on 64-bit SUSE Linux Enterprise Server 11 SP1. In addition to bug fixes and performance improvements, it includes the following changes and enhancements:
The Enterprise Volume Management System (EVMS) has been deprecated in SLES 11, and is also deprecated in OES 11. Novell Linux Volume Manager (NLVM) replaces EVMS for managing NetWare partitions under Novell Storage Services (NSS) pools.
Novell Cluster Services for OES 11 supports NSS pools that are created on OES 11, OES 2 SP3 and earlier, and NetWare 6.5 SP8.
A new NSS capability supports the GPT partitioning scheme. This allows you to create NSS pools up to 8 TB (terabytes) in size on a single device. Pools created with GPT-partitioned devices are not backward compatible with prior releases of OES and NetWare. The DOS partitioning scheme is also available and supports devices up to 2 TB in size.
The NSS management tools use the Novell Linux Volume Manager instead of the Enterprise Volume Management System that is used in previous OES releases. The Storage plug-in to iManager and NSSMU can be used to create pool cluster resources. You can also use NLVM commands to create shared pools and volumes at a command prompt or in scripts.
During a rolling cluster upgrade, the existing NSS pool cluster resources can be cluster migrated to any node in the mixed-node cluster. However, you must not create new NSS pools on OES 11 nodes while you are upgrading the cluster from OES 2 to OES 11. For information, see NSS Pools
in the OES 11: Novell Cluster Services for Linux Administration Guide.
Novell Cluster Services uses a Linux Logical Volume Manager (LVM) volume group and logical volume to create the cluster resource for shared Linux POSIX file systems (such as Ext2, Ext3, ReiserFS, and XFS) on OES 11.
The NSS Management Utility (NSSMU) and Novell Linux Volume Manager (NLVM) commands support creating Linux POSIX file systems and Linux Logical Volume Manager (LVM) volume groups and logical volumes. The tools support both the DOS and the GPT partitioning schemes. The DOS partitioning scheme supports devices up to 2 TB in size. The GPT partitioning scheme supports devices up to 8 zettabytes (ZB, or one billion terabytes). Your actual device size is limited by your storage hardware and the size recognized by your target file system. For information about maximum file system sizes on Linux, see the SUSE Linux Enterprise Server Technical Information: File System Support.
You can create a Linux volume group cluster resource by using NSSMU and Novell Linux Volume Manager commands. You can also use native Linux LVM2 commands to create a shared LVM volume group, and then create a resource by using the generic file system (Generic_FS) resource template in iManager, or by using other application resource templates that need shared Linux POSIX file systems. For information, see Configuring and Managing Cluster Resources for an LVM volume groups
in the OES 11: Novell Cluster Services for Linux Administration Guide.
On OES 2, Linux POSIX volume cluster resources use a Cluster Segment Manager (CSM) container on devices that are managed by EVMS. Because EVMS has been deprecated in OES 11, you must modify the scripts and cluster settings so that the CSM containers can run on OES 11 clusters. You cannot create new cluster resources with CSM containers on OES 11 clusters. For information, see Upgrading and Managing Cluster Resources for Linux POSIX Volumes with CSM Containers
in the OES 11: Novell Cluster Services for Linux Administration Guide.
The Cluster Segment Manager Import/Export (CSMPORT, /opt/novell/ncs/bin/csmport) utility allows you to import and use Linux POSIX volume cluster resources that use CSM containers in OES 11 clusters. After the resource is configured to run on OES 11, you must configure its Preferred Nodes list to ensure that it can fail over only to OES 11 nodes in the mixed-node cluster. For information about CSMPORT, see Cluster Segment Manager Import/Export (csmport) Utility
in the OES 11: Novell Cluster Services for Linux Administration Guide.
The following cluster resource templates have been modified to use LVM volume groups and logical volumes for cluster resources that share data on Linux POSIX file systems. Previously, the templates used the EVMS Cluster Segment Manager container and Linux POSIX volumes.
|
OES 11 Application |
Cluster Resource Template |
|---|---|
|
Archive and Version Services |
AV_Template |
|
DHCP |
DHCP_Template (for an NSS pool or for an LVM volume group) |
|
Linux POSIX file system |
Generic_FS_Template |
|
iFolder |
iFolder_Template (for an NSS pool or for an LVM volume group) |
|
iPrint |
IPrint_Template (for an NSS pool or for an LVM volume group) |
|
MySQL 5.x |
MySQL_Template |
|
Samba |
Samba_Template |
|
Xen virtual machine |
Xen_Template |
The DNS_Template uses an NSS file system. The Generic_IP_Service and XenLive_Template templates do not use a shared data location.
The monitor scripts for resources that use a Linux volume group were modified to check the status of the LVM logical volume in addition to the file system and the IP address.
Ext3 is the default file system type used in the scripts. The Ext2, Ext3, ReiserFS, and XFS file systems have been tested and are fully supported.
The default virtual server name for cluster resources now uses hyphens instead of underscores, such as MYCLUS-MYPOOL-SERVER. The suggested name is compliant with the Internet Engineering Task Force (IETF) RFC 1123 standard that allows host names to contain only letters, digits, and hyphens. Underscores can still be used in the virtual server name if your network environment supports them.
Novell CIFS provides a monitor command option in OES 11 that provides a restart capability if the cifsd daemon goes down. If you create a new pool cluster resource with CIFS enabled as an advertising protocol, the following line is added to the resource’s monitor script:
exit_on_error rcnovell-cifs monitor
Previously, the CIFS status command was used. You can replace it with the monitor command for existing pool cluster resources to take advantage of the CIFS restart capability. For information, see Configuring a Monitor Script for the Shared NSS Pool
in the OES 11: Novell Cluster Services for Linux Administration Guide.
If you attempt to online or migrate a cluster resource to a node that is not in the resource’s Assigned Nodes list, the resource stays offline or is not migrated. This change makes the command behavior consistent with the online and migrate options in the Cluster plug-in in iManager. The node that you specify must be running in the cluster and must also be in the resource’s Assigned Nodes list.
Previously, if the specified node was not a preferred node, the cluster online and cluster migrate commands brought the resource online on a node in its Assigned Nodes list.
When you configure the cluster node in YaST, the LDAP server list is created. The default order is to list the local LDAP server first and others second. In previous OES releases, the default order was based on the IP address.
You cannot change the order of LDAP servers in the list during the cluster node configuration in YaST, but you can modify it later by running the /opt/novell/ncs/install/ncs_install.py script. For information, see Changing the Administrator Credentials or LDAP Server IP Address for a Cluster
in the OES 11: Novell Cluster Services for Linux Administration Guide.
Novell Cluster Services added the Cascade 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.