Replies: 1 comment 9 replies
-
@cyroxxer Could you please share the output of the following Prometheus query? You can also email the output to us at [email protected].
|
Beta Was this translation helpful? Give feedback.
9 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello, i hope me question is at the right place and i'am able to explain the issue/question.
With last two versions 24.08.0 and 24.05.0 we see an issue with some of the metrics that use aggr data. For this example i will try to explain the behavior with the default Aggregate dashboard and the "Top 25 Aggregates by Physical Space Used" graph. With versions before 24.05.0 we were able to get graphs until the day we startet with Harvest/Grafana (Prom retention set to 1 Year). With the latest dashboards we see some graphs don't go behind a specific date in June. In June we renamed some Aggregates ...
As soon i edit the Query from the "Top 25 Aggregates by Physical Space Used" and remove the TOPK section the graph is able to display the data that we assume to see
aggr_space_physical_used{datacenter=
"$Datacenter",cluster="$Cluster",node="$Node",aggr="$Aggregate"}and
topk($TopResources, avg_over_time(aggr_space_physical_used{datacenter=
"$Datacenter",cluster="$Cluster",node="$Node",aggr="$Aggregate"}[3h] @ end()))Can somebody explain what is happening? Is there an error within the queries or do we need to modify something within our dataset in prometheus?
Thanks for helping
Beta Was this translation helpful? Give feedback.
All reactions