Cmak: Partitions Covered & Negative Lag

[lconnell]

Can someone explain why KM reports partitions covered as 0%? Also, why would there be a negative lag?

kafka-manager: 1.3.3.22
kafka: 2.0.1
zookeeper: 3.4.9

Thanks!

[lconnell]

Nobody? Even if it's a pretentious comment pointing me to documentation that I missed? :)

[danielmotaleite]

The negative lag is probably a CMAK bug, where data is fetched from the kafka clusters in different times and in between messages were consumed (ie: get topic offset, then get consumer offset 1 second after). If the fetch order were reversed, the lag would be always positive, but also probably always higher than reality if the messages are quickly consumed.