Add ability to specify replicas for deployment

[awalther52]

For availability reasons, it would be ideal to have more than one replica when running as a deployment. Right now replicas is hardcoded to one. If the node the pod is running on goes down, we loose this metric until the pod is rescheduled onto another node.

Is there a reason why this was not added from the start?

I'd be happy to put up a PR for the helm chart edits!

[jmcgrath207]

Sounds great to me! No reason ATM. I am open to a PR for this with a default of one replica.

If this doesn't work as intended for you, then we will have to look into the leader election at the Golang level.

https://github.com/kubernetes/client-go/blob/master/examples/leader-election/main.go#L122