eG Monitoring
 

Measures reported by AWSMSKFollowTest

Kafka's replication feature provides high availability by optionally persisting each partition on multiple brokers. In a replicated partition, Kafka will write messages to only one replica - the partition leader. The other replicas are followers, which fetch copies of the messages from the leader. Consumers may read from either the partition leader or from a follower.

This test reports the mean time of the request from the follower broker to the leader broker that waits in the queue and the mean time of the response from the leader broker to the follower broker that waits in the queue. Using this test, administrators can proactively figure out if the mean time for both the request and response queues takes too long to process.

Outputs of the test: One set of results for the target AWS Managed Service Kafka server.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
fetchFollowerLocalTime Indicates the mean time that the follower request is processed at the leader. Seconds

 

fetchFollowerRqstQueTime Indicates the mean time that the follower request waits in the request queue. Seconds

 

fetchFollowerRespQueTime Indicates the mean time that the follower request waits in the response queue. Seconds

 

fetchFollowerRespSndTime Indicates the mean time for the follower to send a response. Seconds

 

fetchFollowerTotalTime Indicates the mean total time that followers spend on fetching data from the broker. Seconds