Currently, the chart deploys 2 different collectors, one to scrape the cluster resources as daemonset and other to retrieve k8s api information as deployment with a single replica.

In latest versions, both receivers (k8s_cluster and k8sobjects) support a parameter to enable leader election k8s_leader_elector, which improves the reliability of the receiver adding HA capabilities.

My proposal is unifying both collectors in a single one, running as daemonset with k8s_leader_elector to keep a single instance pulling that info from k8s api.

I'm willing to open a PR with the needed changes, but I'm not fully sure if this will be accepted as removing the second collector could be a breaking change. If you think that this is worth, I'll open a PR later on

In case of not agreeing with unifying them, I'd like to as to at least, support multiple instances of cluster collector to have it with HA config just adding configurable instances to that collector + leader election