5.8 GroupWise Cluster

5.8.1 Installing Reload in a GroupWise Cluster

Reload can be implemented where GroupWise is in a cluster.

When GroupWise is implemented in a cluster there are a few extra steps needed to configure Reload to operate successfully.

In this example there are three servers:

  • Reload Server

  • GroupWise Cluster Node 1 (NODE1)

  • GroupWise Cluster Node 2 (NODE2)

5.8.2 Installation Collector/Server Model

On the Reload Server

Install Reload on the Reload server

Create a Post Office Profile for the Post Office on the GroupWise Cluster

If upgrading a standard server model profile to a Collector/Server model profile:

In Reload Web Administration, choose the profile and go to the Configure tab for the profile.

Select the Post Office and POA Settings panel

Find the line Note: Disk Space Required for Collector

Check on each node that there is sufficient space for the collector and data on each node. Note: This calculation does not take Document Management Libraries into account. If you wish to add DMS, multiple the size of the GWDMS directory structure by 2.

Set "Allow Paired Collector" to Enabled

On GroupWise Cluster Node 1 (NODE1)

Install the Reload Collector Software on NODE1

Browse to the Reload Server and select the appropriate collector software from the Tools dropdown menu and select go, or copy the software from the Reload server found in the directory:

/opt/beginfinite/reload/setup/collector/reload_collector.rpm

Run the following command from the terminal:

rpm -ivh reload_collector.rpm

Exchange SSH keys with the Reload Server

Run the following command from the terminal:

ssh-copy-id -i /root/.ssh/id_rsa.pub <address of the Reload Server>

This is to setup the ssh connection so NODE1 can rsync with Reload without being prompted for a password. This must be successful before proceeding.

Create a Reload Collector Profile

  • Run the following command from the terminal:

    reload

  • Choose Create Profile

  • Select "Create (Create Profile)"

  • Select "Post Office (Create a New GroupWise Post Office Profile)"

  • Give the Profile a Name. The profile name should not exceed eight characters. Use letters and numbers only, no spaces or other characters. For example: po. Be aware, that a profile cannot be renamed, once it has been created. Choose the name wisely. Perhaps you will want to name the profile so that its name is the same as the GroupWise post office or domain that the profile represents. Select the Next button.

  • Give the Profile a Description. The profile description should not exceed 60 characters and should not have comma(,), nor dollar sign($) characters. For example: Domain Collector. The description of the profile will be added to the Agent Log when backup jobs are run for the profile. Select the Next button.

  • Enter the path or browse to the local destination directory to store the backup data. For example: /reload. This path can be a location on the Linux server, or an NFS mount to another Linux/Unix server, or to a SAN. The path should always be available; Reload is not configured to mount paths in order to access stored data. Also, the path should be to a location that has a lot of disk space available. Select the Next button.

  • Confirm the path to the local destination directory

  • Choose the server type: "Collector (A Local Post Office/Domain Paired to a Reload Server)". Select the Next button.

  • Enter the TCP/IP Address or DNS hostname for the Reload server this collector will be paired with. Select the Next button.

  • A terminal session will appear to prompt you to accept an SSH key and enter the password for the Reload server.

  • Enter the SSH port used by the Reload server (default port: 22)

  • Enter the destination path the data will be stored in on the Reload server. This is the destination directory you entered when creating the Paired Server on the Reload server.

  • Enter the source path to the local destination directory entered earlier on the GroupWise server. For example: /groupwise/po

  • Confirm the path to the local source directory

  • A summary of the profile settings will appear. Select the OK button.

  • Test and Initialization:

    If the profile is correct, select the Test button.

    If the profile is not correct, select the Skip button.

  • The profile creation wizard tests and initializes the newly created profile and prepares it to accept a pairing request from a Paired Collector and initializes the newly created profile and prepares it to accept a pairing request from a Paired Collector.

Browse to the Reload Web Administration page on NODE1. For example, http://NODE1:5555.

  • Choose the profile being configured

  • Select the Configure tab and open the Reload Collector Settings panel

  • Set "This Collector is on a Cluster Node" to Enabled

  • Select the Backups tabs, open the Reload Job Control panel and click on "Start a Standard Backup Job [SmartPurge Disabled]" to run a backup job, allow the backup to complete normally.

On GroupWise Cluster Node 2 (NODE2)

Profile settings are not shared across the cluster automatically, this needs to be done manually when setting up or when settings are changed.

  • Install the Reload Collector Software on NODE1

    • Browse to the Reload Server and select the appropriate collector software from the Tools dropdown menu and select go, or copy the software from the Reload server found in the directory: /opt/beginfinite/reload/setup/collector/reload_collector.rpm

    • Run the following command from the terminal: rpm -ivh reload_collector.rpm

  • Exchange SSH keys with the Reload Server

    • Run the following command from the terminal: ssh-copy-id -i /root/.ssh/id_rsa.pub <address of the Reload Server>

    • This is to setup the ssh connection so NODE1 can rsync with Reload without being prompted for a password. This must be successful before proceeding.

  • Obtain the following files from NODE1 and copy them to the same directories on NODE2

    • The contents of the /opt/beginfinite/reload/conf directory (including the ../conf/cron directory)

    • The license file /opt/beginfinite/reload/license/reload.pem

  • Restart the Reload daemon on NODE2 with the following command: reloadd restart

  • Browse to the Reload Web Administration page on NODE2. For example, http://NODE2:5555.

    • Select the Backups tabs, open the Reload Job Control panel and click on "Start a Standard Backup Job [SmartPurge Disabled]" to run a backup job.

    • This backup should '''fail'''

In the Event Log the failure will look like:

 DATE: Fri_Dec_11 TIME: 14:57:25 PROFILE: COLLECT1 - STARTING JOB FOR PROFILE: COLLECT1

 DATE: Fri_Dec_11 TIME: 14:57:25 PROFILE: COLLECT1 - DISK SPACE IN USE AT JOB BEGINNING: 46%

 DATE: Fri_Dec_11 TIME: 14:57:25 PROFILE: COLLECT1 - JOB TYPE: COLLECTOR (INCREMENTAL) BACKUP

 DATE: Fri_Dec_11 TIME: 14:57:25 PROFILE: COLLECT1 - [GRE_DBS_AGENTS] Start Database Backup

 DATE: Fri_Dec_11 TIME: 14:57:34 PROFILE: COLLECT1 - [GRE_DBS_OFUSER] Start OFUSER Backup

 DATE: Fri_Dec_11 TIME: 14:57:38 PROFILE: COLLECT1 - [GRE_DBS] Terminating Processing, Post Office Volume for Profile Not Mounted.

 DATE: Fri_Dec_11 TIME: 14:57:38 PROFILE: COLLECT1 - [GRE_DBS] This must not be the active node for the Post Office

On NODE1 & NODE2

Only the Reload server requires Backup Age monitoring

  • Disable Backup Age Monitoring

    • Browse to the Reload Web Administration page on each node

    • Go to the Collector Profile, under the Configure tab, open the Preferences panel

      • Under Backup Age Monitoring, set "Monitor How Old the Most Current Backup Is" to Disabled

5.8.3 Upgrading Collectors in a Cluster

Clusters are a bit of a challenge to work with as an inactive node does not do anything since it is inactive, therefore you will have to upgrade each node manually.

  • Browse to the Reload Server and select the appropriate collector software from the Tools dropdown menu and select Go, or copy the software from the Reload server found in the directory: /opt/beginfinite/reload/setup/collector/reload_collector.rpm

  • Place the file on each '''inactive''' node in: /opt/beginfinite/reload/upgrade

  • Upgrade the Reload Collector software from the terminal by running the command: reloadu