CPU and RAM Usage in Kubernetes

-- cpu_usage
  sum by (instance) (rate(container_cpu_usage_seconds_total{id="/"}[5m])) / 
  sum by (instance) (machine_cpu_cores{}) * 100
-- cpu usage (GCP)
max(kubernetes_io:node_cpu_allocatable_utilization{project_id="..."})
 
-- ram_usage
  sum by (instance) (container_memory_working_set_bytes{id="/"}) / 
  sum by (instance) (machine_memory_bytes{}) * 100
 
sum by (instance_name)(avg_over_time(compute_googleapis_com:instance_memory_balloon_ram_used{monitored_resource="gce_instance", project_id="sit-varnish-lvzt-run-eaab"}[5m]))
/
sum by (instance_name)(avg_over_time(compute_googleapis_com:instance_memory_balloon_ram_size{monitored_resource="gce_instance", project_id="sit-varnish-lvzt-run-eaab"}[5m]))  

Disk Usage

-- Persistent Volumes in Kubernetes
  sum by (instance,namespace,persistentvolumeclaim) (kubelet_volume_stats_used_bytes{}) / 
  sum by (instance,namespace,persistentvolumeclaim) (kubelet_volume_stats_capacity_bytes{}) * 100
 
-- Disk Usage in Kubernetes
  sum by (instance,device) (container_fs_usage_bytes{id="/", device=~"/dev/.*"}) / sum by (instance,device) (container_fs_limit_bytes{id="/", device=~"/dev/.*"}) * 100
 
-- Disk Usage in GKE
sum by (node_name) (kubernetes_io:node_ephemeral_storage_used_bytes{cluster_name="$cluster"}) /
sum by (node_name) (kubernetes_io:node_ephemeral_storage_allocatable_bytes{cluster_name="$cluster"})

GCP Uptime Check

sum(sum_over_time(monitoring_googleapis_com:uptime_check_check_passed{monitored_resource="uptime_url"}[15m])) by (checked_resource_id)
/
sum(count_over_time(monitoring_googleapis_com:uptime_check_check_passed{monitored_resource="uptime_url"}[15m])) by (checked_resource_id)

Error Rate

  sum(rate(request_count{is_error="true"}[15m])) by (service) 
/ sum(rate(request_count{               }[15m])) by (service) 

Latency

histogram_quantile(0.95, sum(rate(response_time_bucket{}[60m])) by (service, le))

RabbitMQ

-- 0 if queue is empty, 1 - otherwize
(rabbitmq_queue_messages{vhost="/", queue="queue_name"} > bool 0)