Skip to content

Openshift / API Server

Dashboard shows extended statistics about Openshift API server

Tags

  • apiserver
  • openshift

Panels

Overview

Name Description Thresholds Repeat
API server status Shows API server status depending on the number of endpoints exposing metrics Default:
Mode: absolute
Level 1: 1
Level 2: 2

API Servers Shows the number of running API servers depending on the number of endpoints exposing metrics Default:
Mode: absolute
Level 1: 1
Level 2: 2

Current Error Rate Show the current API server error rate: percentage of requests with errors out of the total number of requests Default:
Mode: absolute
Level 1: 3

Average API Server Latency Shows the average duration of request to API server Default:
Mode: absolute
Level 1: 0.2

Average Etcd Requests Latency Shows the average duration of requests to Etcd Default:
Mode: absolute
Level 1: 0.2

Current Controller work queue Depth Show the current depth of controller work queue Default:
Mode: absolute
Level 1: 5

API server nodes status Shows API server nodes status
Request Rate Shows the number of requests per second by each API server instance Default:
Mode: absolute
Level 1: 80

Requests

Name Description Thresholds Repeat
Request rate by resource and verbs Shows the number of requests per second by verb and resource Default:
Mode: absolute
Level 1: 80

Request Rate by resource and group Shows the number of requests per second by resource and its API group in format: API/resource
Request Read and Write Shows rate of request separated by read and write Default:
Mode: absolute
Level 1: 80

Latency

Name Description Thresholds Repeat
Request duration 0.99 quantile Shows the request duration 0.99 quantile: 99% of requests has equal or less duration Default:
Mode: absolute
Level 1: 80

Request duration by verb 0.99 quantile Shows the request duration 0.99 quantile by verb: 99% of requests has equal or less duration Default:
Mode: absolute
Level 1: 80

Request duration by resource and group 0.99 quantile Shows the request duration 0.99 quantile by resource type and its API group: 99% of requests . Format: API/resource
Default:
Mode: absolute
Level 1: 80

Errors

Name Description Thresholds Repeat
API Server error percentage Shows the percentage of requests failed with 5xx error by each API server instance Default:
Mode: absolute
Level 1: 80

Error request rate by verb Show the number of requests per second failed with 5xx or 4xx code grouped by verb Default:
Mode: absolute
Level 1: 80

Error request rate by code Show the number of requests per second failed with 5xx or 4xx code grouped by code Default:
Mode: absolute
Level 1: 80

Controller Work Queue

Name Description Thresholds Repeat
Adds to work queue per second Shows the number of adds to controller work queue per second Default:
Mode: absolute
Level 1: 80

Work queue depth Shows the current depth of controller work queue. It should be near 0 Default:
Mode: absolute
Level 1: 80

Item processing time 0.99 quantile Shows 0.99 duration quantile of the item processed from work queue: 99% of items has equal or less processing time Default:
Mode: absolute
Level 1: 80

Item queue time 0.99 quantile Shows 0.99 duration quantile of the item stay work queue: 99% of items has equal or less staying time Default:
Mode: absolute
Level 1: 80

Unfinished work time Show time of unfinished item processing. Large values indicate stuck threads Default:
Mode: absolute
Level 1: 80

Longest running processor time Show longest time of item processing Default:
Mode: absolute
Level 1: 80

Etcd

Name Description Thresholds Repeat
Etcd request rate Shows the number of requests per second sent to Etcd Default:
Mode: absolute
Level 1: 80

Etcd request duration 0.99 quantile Shows the Etcd request duration 0.99 quantile: 99% of requests has equal or less duration Default:
Mode: absolute
Level 1: 80

Request duration 0.99 quantile by operation Shows the Etcd request duration 0.99 quantile by operation: 99% of requests has equal or less duration Default:
Mode: absolute
Level 1: 80

Request duration 0.99 quantile by object type Shows the Etcd request duration 0.99 quantile by object type: 99% of requests has equal or less duration Default:
Mode: absolute
Level 1: 80

Objects stored in etcd Shows the number of objects stored in Etcd Default:
Mode: absolute
Level 1: 80

Go stats

Name Description Thresholds Repeat
CPU usage Shows the CPU usage by each API server instance Default:
Mode: absolute
Level 1: 80

Goroutines Shows the number of goroutines by each API server instance Default:
Mode: absolute
Level 1: 80

Resident memory Show the resident memory usage by each API server instance Default:
Mode: absolute
Level 1: 80

Allocated memory Shows the allocated memory by each API server instance Default:
Mode: absolute
Level 1: 80

Allocations per second Shows the memory allocations per second by each instance Default:
Mode: absolute
Level 1: 80