Hello,
since upgrading to Dremio 25.1.0 (from Dremio 24.x) we are getting a PrometheusDuplicateTimestamps error from our Prometheus/monitoring stack. I’m pretty sure I saw this problem with a 25.0.0 test installation too.
We installed Dremio using this Helm Chart.
The problematic metric seems to be process_start_time_seconds as I get this twice when looking at the /metrics endpoint of the Dremio Master:
$ curl localhost:9096/metrics 2>/dev/null | grep process_start_time_seconds
# HELP process_start_time_seconds Start time of the process since unix epoch in seconds.
# TYPE process_start_time_seconds gauge
process_start_time_seconds 1.72562120382E9
# HELP process_start_time_seconds Start time of the process since unix epoch.
# TYPE process_start_time_seconds gauge
process_start_time_seconds 1.72562120382E9
Am I the only one experiencing this problem?
Thank you in advance and best regards,
Nico