Uploaded image for project: 'Red Hat Advanced Cluster Management'
  1. Red Hat Advanced Cluster Management
  2. ACM-3748

DOC Grafana downsampled data mismatch

XMLWordPrintable

    • Low
    • -
    • No

      1. Proposed title of this DOC update
        1. Grafana downsampled data syncing
      2. What is the nature and description of the request?
        1. When attempting to query historical data if there is a discrepancy between the calculated step value and downsampled data, the result is empty.  For example, if the calculated step value is 5m and the downsampled data is in a 1 hour interval, grafana shows no data. Users are still able to verify data via prometheus query 
      3. Why does the customer need this? (List the business requirements here)
        1. This issue results in problems analyzing historical data
      4. List any affected packages or components.
        1. ACM 2.6+

       

      The cause comes from the fact that for the Thanos Query Frontend to perform additional queries for other downsampling levels when data is missing, a URL query parameter had to be passed.

      Now, the solution: in the configuration of the Query Frontend datasource in Grafana, scroll down to the "Misc" section. Find the `custom query parameters` field and use `max_source_resolution=auto`. document this scenario and teach people to enable the feature, its consequences (like extra queries being executed internally to fulfill a single user-initiated query), etc.

              mdockery@redhat.com Mikela Jackson
              johlong@redhat.com John Long
              Xiang Yin Xiang Yin
              Scott Berens Scott Berens
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

                Created:
                Updated:
                Resolved: