I am using Grafana 4.3.2 with Prometheus 2.0 as a data source. I am trying to display a bar chart of Prometheus on Grafan. The values I retrieve from Prometheus are as follows:
http_request_duration_seconds_bucket{<other_labels>, le="+Inf"} 146
http_request_duration_seconds_bucket{<other_labels>, le="0.005"} 33
http_request_duration_seconds_bucket{<other_labels>, le="0.01"} 61
http_request_duration_seconds_bucket{<other_labels>, le="0.025"} 90
http_request_duration_seconds_bucket{<other_labels>, le="0.05"} 98
http_request_duration_seconds_bucket{<other_labels>, le="0.1"} 108
http_request_duration_seconds_bucket{<other_labels>, le="0.25"} 131
http_request_duration_seconds_bucket{<other_labels>, le="0.5"} 141
http_request_duration_seconds_bucket{<other_labels>, le="1"} 146
http_request_duration_seconds_bucket{<other_labels>, le="10"} 146
http_request_duration_seconds_bucket{<other_labels>, le="2.5"} 146
http_request_duration_seconds_bucket{<other_labels>, le="5"} 146
So, I expect to see 12 buckets, with the values indicated on the right. However, Grafana shows completely different values, as you can see below:

Is there something I'm missing or is Grafana just not supporting Prometheus histograms (resets the “le” label)?
source
share