During OES installation, select Archive and Version Services and Novell Cluster Services options. For Archive and Version Services, select theoption. You must install Novell Cluster Services (NCS) for Linux on the OES 2 Linux server before configuring Archive and Version Services to work in a cluster environment. NCS provides a Archive and Version Services (AV) resource template, which facilitates configuring the service in a cluster environment.
For highly availability of Archive service, you need to enable cluster service on Archive volume and PostgreSQL database. The Archive volume needs to be on an NSS file system and PostgreSQL database directories on Linux POSIX Volumes.
After you have installed, set up, and configured Novell Cluster Services, you need to create NSS and Linux POSIX shared volumes before configuring Archive and Version service to run in a cluster environment.
Archive and Version Services uses PostgreSQL database. Make sure all nodes in the cluster are configured with either all 32-bit architecture machines or all 64-bit architecture machines. Limitations of the database prevent Archive Version and Services from running in a cluster setup with mixed architectures.
To make the Archive volume highly available, you must create create a shared NSS pool and volume on a shared device that can be migrated or failed over to other nodes in a cluster. The shared volume is the location you want to use for the Archive volume.
You must create a shared disk partition for an Primary Volume, if you need to make the Primary Volume highly available.
Create a NSS shared disk partition for an Primary Volume and cluster enable it. For more information, see OES 2 SP3: Novell Cluster Services 1.8.8 Administration Guide for Linux.
You must create a NSS shared disk partition for an Archive Volume and cluster enable it. For more information, see
NOTE:Cluster resource load and unload scripts are automatically generated for pools when they are cluster-enabled.
Using iManager, proceed with the following steps to make the Archive volume highly available.
This procedure uses the following sample values for the virtual cluster server name of the cluster resource you created for the NSS file system, the server context, a LUM group for the users who want to restore file versions, and a username. Make sure to change these values to the actual values in your solution.
Virtual Cluster Server Name: Archive-server
Server Context: finance
LUM group: Archivegp
Click thetask. This opens the Create Unix Workstation Object window.
In thefield, specify Archive-server.
In the finance.field, search and select
In thefield, search and select .
Click Next to create a Unix workstation.
A UNIX WorkStation-Archive-server.finance workstation is created for Primary Volume.
The user who needs to restore the file versions needs to be LUM-enabled. To do this, click thetask.
In the Carlos and click .window, select
Every Linux user must belong to a primary group. In the novlxtier object and click Next.window, select
IMPORTANT:A user must be a part of the novlxtier group to use the NSS File Version Utility for restoring versions.
In the Select Workstations window, select UNIX WorkStation-Archive-server.finance and click Next.
The user Carlos is LUM-enabled and his primary group is novlxtier. Carlos is also given access to UNIX WorkStation-Archive-server.finance workstation.
NOTE:In addition to novlxtier group, user can be added to another groups using iManager. To add a user to Archivegp group, click task. Select the user Carlos and click . Select the tab and add Archivegp group.
The user Carlos is now a member of novlxtier and Archivegp groups.
The PostgreSQL database’s data directories need to be on Linux POSIX file system volumes. Create and mount the Linux POSIX Volumes. For more information, see OES 2 SP3: Novell Cluster Services 1.8.8 Administration Guide for Linux.
On mounting the Linux POSIX Volumes, continue with Section 10.3.3, Configuring Archive and Version Service on the Shared Disk.
NOTE:You must cluster enable the Linux POSIX Volumes, see Section 10.4.2, Creating an Archive Service Cluster Resource for more information.
To configure Archive service to use shared storage, you need to create a shared directory (file system or disk) on the shared disk system and create mount points to that shared file system on each cluster server that will run Archive service.
Edit the arkdatadir.conf file located in /etc/opt/novell/arkmanager/conf/ and change the existing path from /var/opt/novell/arkmanager/data to MOUNT_POINT/novell/arkmanager/data.
Where DATA_PATH points to location of Linux POSIX volumes on a shared disk and MOUNT_POINT points to mounted path of the Linux POSIX volumes on a shared disk. The value of MOUNT_POINT is same as used in the Archive load script. For example, MOUNT_POINT = /mnt/ark
At the command prompt, execute the following command:
Authenticate with admin password. Enter the inputs for the fields database port, username and password, refer help on the left pane of the screen for details on the fields.
To configure Archive and Version Service, refer Section 5.2, Installing and Configuring an Archive Server.