Did you configure the built-in high availability for BES 5.0? If so, did you make any changes to the failover configuration (health scores)? If so, what did you change and why? If you don't mind sharing your changes/configuration, even if every company is not the same and does not share the same priorities and policies, I think this could be a great addition to the community.

__________________
BCSA (4.1, 5.0) | BCSD (4.1, 5.0)

The views expressed by me on Port3101.org are my own and do not necessarily reflect the views of my employer.

Hey Otto, I have implemented HA w/ 5.0.2. I have left the defaults as is. I have been looking for more information other than RIM documentation. Trying to find what Admins are doing in the real world...if anyone has real world experience out there it would be greatly appreciated.

Configuring health parameters for automatic server failover will depend on a couple other things...

- Are MDS-CS & Collaboration (if you're using this) pools created and automatic connection failover enabled?

- Are Attachment Service pools created?

- Is Database mirroring configured?

When I configure the health parameters, I don't have any health parameters between the promotion and failover threshold lines. The default server failover settings don't make much sense. With default settings, between the promotion and failover thresholds is the Access to web/application content health parameter which is the MDS-CS. For example, if the MDS-CS component on the primary BES fails and everything above the promotion threshold line is in a healthy state, a failover occurs but because the Access to web/application content health parameter is below the promotion threshold line, the MDS-CS component on the HA server is not checked so it may not even be running. So failing over to the HA may not have resolved the issue that caused the failover in the first place. So when configuring the promotion/failover threshold lines, don't have any health parameters between the two red lines.

If you created MDS-CS and Collaboration pools and enabled automatic connection failover, you can leave the MDS-CS and Collaboration health parameters below the promotion/failover thresholds. With the pools, if the primary component fails, it'll automatically failover to the standby instance. MDS-CS has two health parameters - Access to web/application content and Push application access. If you're still at defaults, just move the Access to web/application content health parameter below the failover threshold line. If the MDS-CS/Collaboration pools aren't created and you consider these functions to be vital in your environment, move the health parameters above the promotion/failover threshold lines.

If you created your attachment pool(s), leave the Attachment Viewing health parameter below the promotion/failover threshold lines. If you don't have the attachment pool created and you consider attachment viewing to be critical in your environment, move the attachment viewing health parameter above the promotion/failover threshold lines.

If you don't have database mirroring configured in SQL, move the Connection to BlackBerry Configuration Database health parameter above the promotion/failover threshold lines. Even if you have database mirroring configured, you may still want to move this above the threshold lines. Just depends on Admin preference.

Two health parameters use percentages - User accounts and Connection to messaging servers. By default, the percentage is set to 75%. This can't be changed inside the BAS. It can only be changed using the trait tool. I rarely change these percentages.

Hope this helps.

@bb_paws....good info, better then what RIM provides