cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 8

HA with an Additional Poller

Jump to solution

I am working on a project for a customer who is implementing a large Orion environment. They are wanting to implement HA for their main poller so they can bring the website back up in a different datacenter. Currently their virtual plant does not have enough resources to build an exact duplicate of the main poller in the DR datacenter; however, there is an additional poller in that datacenter. Is it possible to put the primary poller and this additional poller in an HA pool and the additional poller still be able to poll nodes assigned to it, while being in standby for the pool?

0 Kudos
1 Solution
Level 13

negative. The HA option only work to pair the main Orion server with a passive Orion server. Or pairing a polling engine with a passive polling engine. In either scenario only the active side of the HA pair/pool can poll.

View solution in original post

0 Kudos
6 Replies
Level 12
Product Manager
Product Manager

HA has no requirement that resources match between members of the pool. You cannot however failover from an Main Poller to an Additional Poller. You may want to consider having a less powerful system at the DR location. In the event of a failure the worst that could occur is that polling is somewhat delayed and the website is slow. In the event of a disaster though, it's better to have something rather than nothing.

Level 15

The currently  Orion environment. have requisits for user HA ????

Remeber need same subnet.

If not I Think you user FoE is better.

0 Kudos
Level 13

negative. The HA option only work to pair the main Orion server with a passive Orion server. Or pairing a polling engine with a passive polling engine. In either scenario only the active side of the HA pair/pool can poll.

View solution in original post

0 Kudos

Is running an Active/Active possible with the HA pool or is this something that would have to be achieved through a NLB?

0 Kudos

Today, HA is only Active/Passive. What are the recovery time objectives (RTO) with the DR plan you're working with? Theoretically, HA should take only a few minutes to fail over, but it only covers the application server, not the back end DB. For the DB side, you'd have to leverage one of Microsoft's HA strategies.

There is also just a cold backup of just the database, but depending on how many Orion modules we're working with it could take several hours to re-install Orion and restore the DB to get it working again.

0 Kudos