Description
We need to provide cluster resource as metrics.
I propose the following metrics for cluster resources.
- tm.cluster.uptime (seconds after startup)
- tm.cluster.nodes.total-num
- tm.cluster.nodes.active-num
- tm.cluster.nodes.lost-num
- tm.cluster.memory.total (total memory in cluster)
- tm.cluster.memory.used (used memory in cluster)
- tm.cluster.cpu.total (total virtual cores in cluster)
- tm.cluster.cpu.used (used virtual cores in cluster)