docs: Issue with "Histograms and summaries" #3068

Activity

1. machugram commented on Oct 20, 2026

   

   machugram

   on Oct 20, 2026

   More actions

   Hi @colinpaicemq, both queries return nothing because of vector matching: the left side keeps an le label and the right side has none, so the division finds no matching series. Ignoring le fixes it:

   sum by (le) (rate(traces_span_metrics_duration_milliseconds_bucket[5m]))
   / ignoring(le) group_left
   sum(rate(traces_span_metrics_duration_milliseconds_count[5m])) * 100
   

   (The second query is also missing a } after le="+Inf".) Buckets are cumulative, so this is "% of spans ≤ le". For percentiles use histogram_quantile(0.95, sum by (le) (rate(..._bucket[5m]))), and for heatmaps see the "Visualizing histograms" section of this page, which links Grafana/Perses.

   This looks more like a usage question than a docs bug, so https://prometheus.io/community/ is a good place for follow-ups, and I think this can be closed.
2. colinpaicemq commented on Oct 22, 2026

   

   colinpaicemq

   on Oct 22, 2026 via email

   Author

   More actions

   Hi Rexford, Great thanks... I'll take your answer and blog it as my problem and your solution Colin

   …

   On Tue, Oct 20, 2026 at 12:58 PM Rexford Machu ***@***.***> wrote: *machugram* left a comment (prometheus/docs#3068) <#3068 (comment)> Hi @colinpaicemq <https://github.com/colinpaicemq>, both queries return nothing because of vector matching: the left side keeps an le label and the right side has none, so the division finds no matching series. Ignoring le fixes it: sum by (le) (rate(traces_span_metrics_duration_milliseconds_bucket[5m])) / ignoring(le) group_left sum(rate(traces_span_metrics_duration_milliseconds_count[5m])) * 100 (The second query is also missing a } after le="+Inf".) Buckets are cumulative, so this is "% of spans ≤ le". For percentiles use histogram_quantile(0.95, sum by (le) (rate(..._bucket[5m]))), and for heatmaps see the "Visualizing histograms" section of this page, which links Grafana/Perses. This looks more like a usage question than a docs bug, so https://prometheus.io/community/ is a good place for follow-ups, and I think this can be closed. — Reply to this email directly, view it on GitHub <#3068?email_source=notifications&email_token=AJQXLWMMI4AKIS5LXVPQDZ35U5HM3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTMMRZGAZTANRUGA42M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-6290306409>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AJQXLWPUQO2TAMJ52PBOSO35U5HM3AVCNFSNUABEKJSXA33TNF2G64TZHMZDQMRTGM4TENB3JFZXG5LFHM2TCMZTGA2DANJXGGQXMAQ> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/AJQXLWPMLMIHBCMDHLKWIKL5U5HM3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTMMRZGAZTANRUGA42M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJKTGN5XXIZLSL5UW64Y> and Android <https://github.com/notifications/mobile/android/AJQXLWJN2GYSGZAARGNHGGL5U5HM3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTMMRZGAZTANRUGA42M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLTGN5XXIZLSL5QW4ZDSN5UWI>. Download it today! You are receiving this because you were mentioned.Message ID: ***@***.***>