Troubleshooting

Description

The Troubleshooting section of the module is intended to provide data capable of assisting with identifying potential points of failure on the cluster.

Displayed Data

• metric charts for failed requests
  • failed requests from producers for every broker per second
  • failed sampling requests for every broker per second
• metric charts for general cluster load
  • overall request rates for every broker per second
  • consumer request rates per every broker per second
  • controller request rates per every broker per second (controller and broker-side)
• metric charts for operations in queue
  • sampling requests waiting in queue
  • producer requests waiting in queue
  • shrink and expansion rates of in-sync replicas per second
• general metric charts
  • rate and time of log flush operations in ms
  • leader election rates on the cluster

Dashboard List

• Failed Requests from Producers per Broker, rqtps
• Failed Sampling Requests per Broker, rqtps
• Total Requests per Broker
• Consumer Requests per Broker, rqtps
• Follower Requests per Broker, rqtps (Broker Side)
• Follower Requests per Broker, rqtps (Controller Side)
• Requests Waiting in Sampling Queue
• Requests Waiting in Producer Queue
• ISR Operations per Second, opps
• Log Flush Stats
• Leader Election Rate