Multi host upgrades

This article provides steps for upgrading multi-host TeamForge Orchestrate installations to the latest release.

Download upgrade_[version].sh

1. To gain access to the TeamForge Orchestrate upgrade script upgrade_[version].sh, contact CollabNet Support at https://www.collab.net/support/secure-customer-login

   where [version] represents the upgrade target version of Orchestrate.

2. Once Support has given you access, download upgrade_[version].sh from https://ctf.open.collab.net/sf/sfmain/do/viewProject/projects.sfdl
3. Copy upgrade_[version].sh to /tmp on each host in question and change directories to that location.

   cp upgrade_[version].sh /tmp cd /tmp chmod 755 upgrade_[version].sh

Back up the TeamForge Orchestrate system

4. Before upgrading, backup TeamForge Orchestrate system and data, see Backup and Restore

Disconnected media

5. If you are using "disconnected media" to upgrade, follow the steps in Upgrading without internet access. Otherwise continue with the MongoDB section below.
   Important: It is imperative that MongoDB and RabbitMQ be upgraded before the Orchestrate service to avoid potential data loss due to downed services. Please proceed in the order prescribed below.

Upgrade MongoDB - Before upgrading MongoDB, to prevent unintentional data loss, it is important that you shutdown the services running on the App server.

6. Log into the host running the TeamForge Orchestrate application (App server) with a privileged account.
7. Shutdown the "orchestrate" service, see System Administration.

   sudo service orchestrate stop

   Note: The TeamForge Orchestrate web interface will be unavailable to end-users until after the upgrade process. RabbitMQ will remain up though, collecting and queuing data from sources.
8. Log into the host running MongoDB (DB server) with a privileged account.
9. Perform upgrade by running upgrade_[version].sh, passing the mongodb flag.

   sudo ./upgrade_[version].sh mongodb

   Upon success, MongoDB will be automatically restarted.

10. You may restart the Orchestrate services on the App server at this time, or continue to the RabbitMQ section.

    sudo service orchestrate start

Upgrade RabbitMQ

11. Log into the host running RabbitMQ (MQ server) with a privileged account.
12. Perform upgrade by running upgrade_[version].sh, passing the rabbitmq flag.

    sudo ./upgrade_[version].sh rabbitmq

    Upon success, RabbitMQ will be automatically restarted.

Upgrade TeamForge Orchestrate

13. Log into the host running the TeamForge Orchestrate application (App server) with a privileged account.
14. Perform upgrade by running upgrade_[version].sh, passing the orchestrate flag.

    sudo ./upgrade_[version].sh orchestrate

    Upon success, Orchestrate services will be restarted automatically.

15. Verify that all services are running after upgrade_[version].sh terminates

    The upgrade script should display a list of all services with [OK] denoting a successful restart.

Upgrade the Collabnet TeamForge tracker adapter - The TeamForge tracker adapter sends tracker changes to Orchestrate and may be dependent on a specific version of TeamForge. To install the tracker adapter:

16. Download the TeamForge Tracker adapter orchestrate-tracker-adapter-[version].jar corresponding to your version of CollabNet TeamForge from the TeamForge Orchestrate Community.
17. Log into your Collabnet TeamForge as a site admin.
18. Navigate to Admin > System Tools > Custom Event Handlers
    Note: If Custom Event Handlers is not available on your instance of TeamForge, please see TeamForge adapters - alternative installation method.
19. Click Create.
20. Browse to the adapter and click Add.

Restart Orchestrate and TeamForge

Once the tracker adapter is installed, a reboot of both TeamForge and Orchestrate is necessary.

21. Restart TeamForge

    See TeamForge documentation for details on restarting TeamForge.

22. Restart Orchestrate

    Restart the "collabnet-nginx" service. See System Administration for details on restarting collabnet-nginx.