What do these Reaper Metrics Mean?

84 views
Skip to first unread message

FH

unread,
Feb 17, 2020, 9:31:26 AM2/17/20
to TLP Apache Cassandra Reaper users
Hi all ...
After integrating reaper with datadog, I can see over 600 metrics display on the admin page (depending on how man clusters are registered). The most useful I have found are ...

  • RepairRunner metrics (3): The repair UUID is embedded in the metric name. Available per cluster & per keyspace.
    • io_cassandrareaper_service_RepairRunner_repairProgress
    • io_cassandrareaper_service_RepairRunner_segmentsDone
    • io_cassandrareaper_service_RepairRunner_segmentsTotal
  • SegmentRunner metrics (3): No repair UUID (uses NULL). Available per cluster & per node
    • # TYPE io_cassandrareaper_service_SegmentRunner_postpone_null_* gauge
    • # TYPE io_cassandrareaper_service_SegmentRunner_repairing_null_* summary
    • # TYPE io_cassandrareaper_service_SegmentRunner_runRepair_null_* summary
  1. Can some one, please, explain what the segmentrunnder metrics mean? I was not able to decipher their meaning by oberving their behavior during a repair.
  2. As for the repairrunner metrics, because the repair run UUID changes with every run, it is not sustainable to keep editing dashboards to remove old UUID metrics and insert new ones. How are others managing this? 

Thanks 

Fd Habash

unread,
Feb 21, 2020, 10:20:56 AM2/21/20
to TLP Apache Cassandra Reaper users
Hi all ...
Where else can I find more information about these metrics? 
Thanks 
Reply all
Reply to author
Forward
0 new messages