How can I query CPU utilization / Memory utilization & no. of pods running only for Worker Nodes Not for cluster

12 views
Skip to first unread message

Manish Gupta

unread,
Jul 15, 2020, 9:10:59 AM7/15/20
to Prometheus Users
1. I want to run below query on worker nodes only, so how can i create a filter/label

sum(rate(container_cpu_usage_seconds_total{id="/"}[5m])) / sum(kube_node_status_allocatable_cpu_cores)  * 100


2.  Actually, I want Proactively worker nodes scale up once it hit threshold value.



Thanks in advance!!

Reply all
Reply to author
Forward
0 new messages