Question about how people are using the Docs from Elementary. We’re currently only using it in the DE team, but I’d like to make it available to stakeholders. But I’m concerned that failures that are resolved will prompt questions. eg a unique test fails, we resolve the issue upstream and things pass again. But it still shows as failed in the dashboard. I’m wondering if a feature or something like the equivalent of pagerduty’s ‘resolved’ would be useful. Say • Uniqueness test fails on 5am run, again on 6,7 am run • Team looks into the issue and resolves the issue • Test passes at 8, 9 • User marks today’s failures of Test ‘unique_tablename_…’ as resolved • Dashboard represnts ‘open issue’ failures changes to look ‘green’

Hi @Jon Simpson! This sounds like an interesting topic. Would you be willing to schedule a call and discuss this further with the product team? You can schedule a call here.

Jumping in to address the question (although I highly recommend you talk with @Hadar Sagiv as well) - The report is a static file, it presents the data relevant to when it was generated. What you could do to keep it relevant is to generate it more frequently, especially to sync the report generation with when you execute

dbt test

command. Also, the report includes a timestamp, so you could maybe instruct to disregard the data if the report is over X hours old.

One last thing to add - the new dashboard we launched this week shows the latest test result of each test, which means it shows the most current issues status (if you generate is frequently as @Maayan Salom suggested). If an issue was resolved and the latest test result was "pass" it will appear as such in the dashboard and you won't see the earlier failure. As mentioned, would love to hear more about how you use Elementary and your collaboration/internal communication needs:)

I mentally forgot that there’s not a backend to retain state when interacting with in the UI so it might not be a great idea. I’ll see if I can find a time with Hadar. Hadar, didn’t realize the results only show the most recent run. That may be sufficient. I’ll give the new version a look early next week