Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Klipfolio data aggregation #22

Closed
eljhkrr opened this issue Apr 8, 2022 · 5 comments
Closed

Klipfolio data aggregation #22

eljhkrr opened this issue Apr 8, 2022 · 5 comments
Assignees

Comments

@eljhkrr
Copy link
Member

eljhkrr commented Apr 8, 2022

Metrics required:

  • Disabled/Enabled data sources
  • Refresh fails

https://docs.google.com/spreadsheets/d/1-BMXlMWrpif3fwI4TN-1VlvJNcdbC4HU1fP3SGIARnQ/edit#gid=0

To be aggregated in #29

@eljhkrr eljhkrr changed the title App Monitoring - Klipfolio datasources Klipfolio data aggregation May 19, 2022
@eljhkrr eljhkrr self-assigned this May 19, 2022
@eljhkrr
Copy link
Member Author

eljhkrr commented Jun 20, 2022

  • Accuracy check
    Muso shows 15 out of 15 data sources failing
    Screenshot from 2022-06-20 00-01-58
    VHW has 37 failing data sources shown as 1%
    Screenshot from 2022-06-19 23-49-40
  • Use stacked bar chart with total data source count, failing data sources & % failing overlay
  • Would be nice to get data on disabled data sources

@eljhkrr
Copy link
Member Author

eljhkrr commented Jul 25, 2022

  • Klipfolio doesn't strictly align datasource refresh with the time window setting. Currently the widget obtains data by comparing last refresh date with current time to determine if the widget is up to date. However, this results in false positives as seen below with musoapp datasources:
    Screenshot from 2022-07-25 16-50-54
    Screenshot from 2022-07-25 16-46-40
    This threshold can be standardized to 3 days to improve usability.

  • Stacked bar chart for total datasource count is implemented by stacking metrics on top of each other:
    Screenshot from 2022-07-25 15-59-12
    Reverted to the layout in the point above for a more natural look.

@eljhkrr
Copy link
Member Author

eljhkrr commented Aug 5, 2022

Feedback from last check-in:

  • Introduce a potentially stale datasource metric based on % of refresh interval and a higher % for failing data sources. In the data snippet above, refresh interval of 43200 vs refresh age of 134501 represents 311%. A cut off of 500% would indicate definite failure.

  • Bars can be stacked and justified by getting the difference from total number of datasources

@eljhkrr
Copy link
Member Author

eljhkrr commented Sep 20, 2022

I've added a potential failed category in addition to total & failed indicators. The potential failed is a buffer for incidents where klipfolio doesn't refresh according to the set interval, allowing about to 5 cycles without an update.
This is what it looks like for PIH:
Screenshot from 2022-09-20 13-05-53
Muso:
Screenshot from 2022-09-20 13-06-42
Brac:
Screenshot from 2022-09-20 13-04-59

@eljhkrr eljhkrr closed this as completed Oct 11, 2022
@eljhkrr
Copy link
Member Author

eljhkrr commented Oct 26, 2022

Interesting case in MoH Mali supervisor app where a file-based datasource is set to never refresh and shows as a refresh failure:
image
image
This is a false positive and datasources like these should be filtered out

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant