How to use grafana dashboard with koperator

[cuisongliu]

Problem Statement

I want to monitor kafka on grafana's dashboard. In addition to creating the corresponding serviceMonitor, how can I create the corresponding dashboard configuration? Has it been written?

Proposed Solution

How can I provide an official dashboard configuration, or do I need to write the corresponding dashboard myself?

[bartam1]

Hello!
Here you can find some useful informations
https://banzaicloud.github.io/koperator-docs/docs/monitoring/
https://github.com/banzaicloud/koperator/blob/master/config/samples/kafkacluster-prometheus.yaml

[bartam1]

monitoringConfig:
    jmxImage: "ghcr.io/banzaicloud/jmx-javaagent:0.16.1"
    pathToJar: "/jmx_prometheus_javaagent.jar"
    kafkaJMXExporterConfig: |
      lowercaseOutputName: true
      lowercaseOutputLabelNames: true
      ssl: false
      whitelistObjectNames:
      # kafka.cluster
      - kafka.cluster:type=Partition,name=UnderMinIsr,*
      - kafka.cluster:type=Partition,name=UnderReplicated,*
      # kafka.controller
      - kafka.controller:type=ControllerChannelManager,name=QueueSize,*
      - kafka.controller:type=ControllerChannelManager,name=TotalQueueSize
      - kafka.controller:type=ControllerStats,*
      - kafka.controller:type=KafkaController,*
      # kafka.log
      - kafka.log:type=Log,name=LogEndOffset,*
      - kafka.log:type=Log,name=LogStartOffset,*
      - kafka.log:type=Log,name=Size,*
      - kafka.log:type=LogManager,*
      # kafka.network
      - kafka.network:type=Processor,name=IdlePercent,*
      - kafka.network:type=RequestChannel,name=RequestQueueSize
      - kafka.network:type=RequestChannel,name=ResponseQueueSize,*
      - kafka.network:type=RequestMetrics,name=ErrorsPerSec,*
      - kafka.network:type=RequestMetrics,name=RequestsPerSec,*
      - kafka.network:type=RequestMetrics,name=LocalTimeMs,request=Produce
      - kafka.network:type=RequestMetrics,name=LocalTimeMs,request=FetchConsumer
      - kafka.network:type=RequestMetrics,name=LocalTimeMs,request=FetchFollower
      - kafka.network:type=RequestMetrics,name=RemoteTimeMs,request=Produce
      - kafka.network:type=RequestMetrics,name=RemoteTimeMs,request=FetchConsumer
      - kafka.network:type=RequestMetrics,name=RemoteTimeMs,request=FetchFollower
      - kafka.network:type=RequestMetrics,name=RequestQueueTimeMs,request=Produce
      - kafka.network:type=RequestMetrics,name=RequestQueueTimeMs,request=FetchConsumer
      - kafka.network:type=RequestMetrics,name=RequestQueueTimeMs,request=FetchFollower
      - kafka.network:type=RequestMetrics,name=ResponseQueueTimeMs,request=Produce
      - kafka.network:type=RequestMetrics,name=ResponseQueueTimeMs,request=FetchConsumer
      - kafka.network:type=RequestMetrics,name=ResponseQueueTimeMs,request=FetchFollower
      - kafka.network:type=RequestMetrics,name=ResponseSendTimeMs,request=Produce
      - kafka.network:type=RequestMetrics,name=ResponseSendTimeMs,request=FetchConsumer
      - kafka.network:type=RequestMetrics,name=ResponseSendTimeMs,request=FetchFollower
      - kafka.network:type=RequestMetrics,name=TotalTimeMs,request=Produce
      - kafka.network:type=RequestMetrics,name=TotalTimeMs,request=FetchConsumer
      - kafka.network:type=RequestMetrics,name=TotalTimeMs,request=FetchFollower
      - kafka.network:type=SocketServer,name=NetworkProcessorAvgIdlePercent
      # kafka.server
      - kafka.server:type=BrokerTopicMetrics,name=BytesInPerSec,*
      - kafka.server:type=BrokerTopicMetrics,name=BytesOutPerSec,*
      - kafka.server:type=BrokerTopicMetrics,name=FailedFetchRequestsPerSec
      - kafka.server:type=BrokerTopicMetrics,name=FailedProduceRequestsPerSec
      - kafka.server:type=BrokerTopicMetrics,name=MessagesInPerSec,*
      - kafka.server:type=BrokerTopicMetrics,name=ReassignmentBytesInPerSec
      - kafka.server:type=BrokerTopicMetrics,name=ReassignmentBytesOutPerSec
      - kafka.server:type=BrokerTopicMetrics,name=ReplicationBytesInPerSec
      - kafka.server:type=BrokerTopicMetrics,name=ReplicationBytesOutPerSec
      - kafka.server:type=BrokerTopicMetrics,name=TotalFetchRequestsPerSec,*
      - kafka.server:type=BrokerTopicMetrics,name=TotalProduceRequestsPerSec,*
      - kafka.server:type=BrokerTopicMetrics,name=FetchMessageConversionsPerSec
      - kafka.server:type=BrokerTopicMetrics,name=ProduceMessageConversionsPerSec
      - kafka.server:type=DelayedOperationPurgatory,*
      - kafka.server:type=FetcherLagMetrics,name=ConsumerLag,*
      - kafka.server:type=FetcherStats,name=BytesPerSec,*
      - kafka.server:type=KafkaRequestHandlerPool,*
      - kafka.server:type=KafkaServer,name=BrokerState
      - kafka.server:type=Fetch
      - kafka.server:type=Produce
      - kafka.server:type=Request
      - kafka.server:type=app-info,*
      - kafka.server:type=ReplicaFetcherManager,*
      - kafka.server:type=ReplicaManager,*
      - kafka.server:type=SessionExpireListener,*
      - kafka.server:type=ZooKeeperClientMetrics,name=ZooKeeperRequestLatencyMs
      - kafka.server:type=socket-server-metrics,listener=*,*
      # java
      - java.lang:type=*
      - java.nio:*
      rules:
      # kafka.cluster
      - pattern: kafka.cluster<type=(Partition), name=(UnderMinIsr|UnderReplicated), topic=([-.\w]+), partition=(\d+)><>(Value)
        name: kafka_controller_$1_$2_$5
        type: GAUGE
        labels:
          topic: $3
          partition: $4
        cache: true
      # kafka.controller
      - pattern: kafka.controller<type=(ControllerChannelManager), name=(QueueSize), broker-id=(\d+)><>(Value)
        name: kafka_controller_$1_$2_$4
        type: GAUGE
        labels:
          broker_id: $3
        cache: true
      - pattern: kafka.controller<type=(ControllerChannelManager), name=(TotalQueueSize)><>(Value)
        name: kafka_controller_$1_$2_$3
        type: GAUGE
        cache: true
      - pattern: kafka.controller<type=(ControllerStats), name=([-.\w]+)><>(Count)
        name: kafka_controller_$1_$2_$3
        type: COUNTER
        cache: true
      - pattern: kafka.controller<type=(KafkaController), name=([-.\w]+)><>(Value)
        name: kafka_controller_$1_$2_$3
        type: GAUGE
        cache: true
      # kafka.log
      - pattern: kafka.log<type=Log, name=(LogEndOffset|LogStartOffset|Size), topic=([-.\w]+), partition=(\d+)><>(Value)
        name: kafka_log_$1_$4
        type: GAUGE
        labels:
          topic: $2
          partition: $3
        cache: true
      - pattern: kafka.log<type=(LogManager), name=(LogDirectoryOffline), logDirectory="(.+)"><>(Value)
        name: kafka_log_$1_$2_$4
        labels:
          log_directory: $3
        type: GAUGE
        cache: true
      - pattern: kafka.log<type=(LogManager), name=(OfflineLogDirectoryCount)><>(Value)
        name: kafka_log_$1_$2_$3
        type: GAUGE
        cache: true
      # kafka.network
      - pattern : kafka.network<type=(Processor), name=(IdlePercent), networkProcessor=(\d+)><>(Value)
        name: kafka_network_$1_$2_$4
        labels:
          network_processor: $3
        type: GAUGE
        cache: true
      - pattern: kafka.network<type=(RequestChannel), name=(RequestQueueSize)><>(Value)
        name: kafka_network_$1_$2_$3
        type: GAUGE
        cache: true
      - pattern : kafka.network<type=(RequestChannel), name=(ResponseQueueSize), processor=(\d+)><>(Value)
        name: kafka_network_$1_$2_$4
        type: GAUGE
        labels:
          processor: $3
        cache: true
      - pattern : kafka.network<type=(RequestMetrics), name=(ErrorsPerSec), request=(\w+), error=(\w+)><>(Count)
        name: kafka_network_$1_$2_$5
        type: COUNTER
        labels:
          request: $3
          error: $4
        cache: true
      - pattern : kafka.network<type=(RequestMetrics), name=(RequestsPerSec), request=(\w+), version=(\d+)><>(Count)
        name: kafka_network_$1_$2_$5
        type: COUNTER
        labels:
          request: $3
          version: $4
        cache: true
      - pattern : kafka.network<type=(RequestMetrics), name=(\w+TimeMs), request=(\w+)><>(Count)
        name: kafka_network_$1_$2_$4
        type: COUNTER
        labels:
          request: $3
        cache: true
      - pattern : kafka.network<type=(SocketServer), name=(NetworkProcessorAvgIdlePercent)><>(Value)
        name: kafka_network_$1_$2_$3
        type: GAUGE
        cache: true
      # kafka.server
      - pattern: kafka.server<type=(BrokerTopicMetrics), name=(\w+PerSec), topic=([-.\w]+)><>(Count)
        name: kafka_server_$1_$2_$4
        type: COUNTER
        labels:
          topic: $3
        cache: true
      - pattern: kafka.server<type=(BrokerTopicMetrics), name=(\w+PerSec)><>(Count)
        name: kafka_server_$1_$2_total_$3
        type: COUNTER
        cache: true
      - pattern: kafka.server<type=(DelayedOperationPurgatory), name=(\w+), delayedOperation=(\w+)><>(Value)
        name: kafka_server_$1_$2_$4
        type: GAUGE
        labels:
          delayed_operation: $3
        cache: true
      - pattern: kafka.server<type=(FetcherLagMetrics), name=(ConsumerLag), clientId=([-.\w]+), topic=([-.\w]+), partition=(\d+)><>(Value)
        name: kafka_server_$1_$2_$6
        type: GAUGE
        labels:
          client_id: $3
          topic: $4
          partition: $5
        cache: true
      - pattern: kafka.server<type=(FetcherStats), name=(\w+PerSec), clientId=([-.\w]+), brokerHost=([-.\w]+), brokerPort=(\d+)><>(Count)
        name: kafka_server_$1_$2_$6
        type: COUNTER
        labels:
          client_id: $3
          broker_host: $4
          broker_port: $5
        cache: true
      - pattern: kafka.server<type=(KafkaRequestHandlerPool), name=(\w+)><>(Count)
        name: kafka_server_$1_$2_$3
        type: COUNTER
        cache: true
      - pattern: kafka.server<type=(KafkaRequestHandlerPool), name=(\w+)><>(OneMinuteRate)
        name: kafka_server_$1_$2_$3
        type: GAUGE
        cache: true
      - pattern: kafka.server<type=(KafkaServer), name=(\w+)><>(Value)
        name: kafka_server_$1_$2_$3
        type: GAUGE
        cache: true
      - pattern: kafka.server<type=(Fetch|Produce|Request)><>(queue-size)
        name: kafka_server_$1_$2
        type: GAUGE
        cache: true
      - pattern: kafka.server<type=(app-info), id=(\d+)><>(StartTimeMs)
        name: kafka_server_$1_$3
        type: COUNTER
        labels:
          broker_id: $2
        cache: true
      - pattern: 'kafka.server<type=(app-info), id=(\d+)><>(Version): ([-.~+\w\d]+)'
        name: kafka_server_$1_$3
        type: COUNTER
        labels:
          broker_id: $2
          version: $4
        value: 1.0
        cache: false
      - pattern: kafka.server<type=(ReplicaFetcherManager), name=(\w+), clientId=([-.\w]+)><>(Value)
        name: kafka_server_$1_$2_$4
        type: GAUGE
        labels:
          client_id: $3
        cache: true
      - pattern: kafka.server<type=(ReplicaManager), name=(\w+)><>(Value)
        name: kafka_server_$1_$2_$3
        cache: true
      - pattern: kafka.server<type=(ReplicaManager), name=(\w+)><>(Count)
        name: kafka_server_$1_$2_$3
        type: COUNTER
        cache: true
      - pattern: kafka.server<type=(SessionExpireListener), name=(\w+)><>(Value)
        name: kafka_server_$1_$2_$3
        type: GAUGE
        cache: true
      - pattern: kafka.server<type=(SessionExpireListener), name=(\w+)><>(Count)
        name: kafka_server_$1_$2_$3
        type: COUNTER
        cache: true
      - pattern: kafka.server<type=(ZooKeeperClientMetrics), name=(\w+)><>(\d+)thPercentile
        name: kafka_server_$1_$2
        type: GAUGE
        labels:
          quantile: 0.$3
        cache: true
      - pattern: kafka.server<type=(ZooKeeperClientMetrics), name=(\w+)><>(Mean|Max|StdDev)
        name: kafka_server_$1_$2_$3
        type: GAUGE
        cache: true
      - pattern: kafka.server<type=(socket-server-metrics), listener=([-.\w]+), networkProcessor=(\d+)><>([-\w]+(?:-count|-rate))
        name: kafka_server_$1_$4
        type: GAUGE
        labels:
          listener: $2
          network_processor: $3
        cache: true
      # Export all other java.{lang,nio}* beans using default format
      - pattern: java.lang.+
        cache: true
      - pattern: java.nio.+
        cache: true