Add monitoring endpoint for oldest task in a given status

[benoit74]

Rationale

Add an endpoint which can be called from UptimeRobot to e.g. monitor oldest running task

Changes

• add a generic /status/{monitor_name} endpoint
• implement a first monitor oldest_task_older_than which takes two parameters (as query string):
  • status: a task status which has an associated timestamp, e.g. scraper_started, scraper_completed, cancel_requested
  • threshold_secs: a threshold which cause the monitor to fail if oldest task is older than this threshold, in seconds
  • returns a textual answer with HTTP code 200 and details if monitor succeeded to run:
    • e.g. oldest_task for scraper_started and 500s: OK or oldest_task for scraper_started and 5s: KO
  • returns HTTP code 400 on bad requests
• update (manually 🤢) OpenAPI spec
• add automated tests
  • slightly modify test fixtures so that tasks have timestamp a bit in the past (we didn't cared about their values so far)

The : KO or : OK strings can be searched for in UptimeRobot KW probe. Always returning 200 even in case of "problem" is deemed more appropriate to respect HTTP paradigms and allows to more easily differentiate a buggy API from a down monitor.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.49%. Comparing base (7aaa114) to head (02a4a48).
Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1076      +/-   ##
==========================================
+ Coverage   88.38%   88.49%   +0.11%     
==========================================
  Files          98      101       +3     
  Lines        5482     5538      +56     
==========================================
+ Hits         4845     4901      +56     
  Misses        637      637              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[rgaudin]

Good ; cleanup the commented code first though

[benoit74]

🤦🏻‍♂️thank you