Skip to content

Web app for histogram monitoring

Edward Karl Galantay requested to merge egalanta into master

This is the implementation of the web app for histogram monitoring based on Flask and Plotly.

It's located at daq/scripts/Monitoring/Dashboard.

The web app reads the Redis database and looks for modules with "monitor" in its name and updates the histograms every 3 seconds.

The web app contains:

  • Tag system: every histogram has an ID: "<module_name>-<hist_name>" and the default tags for an histogram are <module_name> and <hist_name>.
  • User tags: users can assign custom tags under the histograms. The histograms can then be searched by user tags and default tags.
  • Live views: the user can choose three different views: by module, by tags, and unique view (if the user wants to look only one histogram, useful if the histogram is very large)
  • Historic view: it is possible to see previous states of the histograms. Two APScheduler jobs will save respectively the histograms every 1 minute and 30 minutes in the Redis database. Only the histograms saved every 30 minutes are permanently stored. Histograms saved every minute are deleted after 30 minutes.
  • Download/Upload tags: the user can download a JSON file with the tags (default+user) as a backup and can re-upload it later
Edited by Edward Karl Galantay

Merge request reports