Cratedb daily IOPS0 spikes

CrateDB
1

Hello Everyone,

We are monitoring one of our production cluster with grafana/prometheus for a data streaming pipeline using kafka/storm and cratedb and noticed a daily IOPS spikes for cratedb, so we are trying to understand whether if this is related to one of the crateDB core tasks like stats.service.interval or udc.interval or other services that could cause a high IOPS.

image
image871×347 15.4 KB

Any thoughts are appreciated.
Thank you

2

Hey there, thanks for flying with CrateDB!

Most likely this is triggered by Cluster-wide settings — CrateDB: Reference.

Throttling of this is already merged https://github.com/crate/crate/pull/13046. Which version are you running?

regards,
Walter

1 Like
3

Thank you for your prompt reply.
We are currently using cratedb 4.6.6