sentry
sentry copied to clipboard
generating backpressure dd events
Description
Now that sentry supports Datadog events,
updating backpressure code to generate events as well (in addition to metrics).
Next steps
- Send events using dogshell
- Update event body if the need be
What is the motivation for this? Does it tell us anything that the metric doesn't already?
Hey @lynnagara there was some more detail in the original PR / ticket: https://getsentry.atlassian.net/browse/OPS-5392
This would enable us to do things like:
- overlay events on top of dashboard widgets
- add tags to events to support analysis / grouping
- Trigger DD monitors for alerting purposes (eg. notifying slack / PagerDuty)
Here's an example time period where we experienced backpressure on May 6:
And split:
Some things that aren't clear:
- Why is everything a multiple of 10?
- When exactly did backpressure trigger?
- When did it resolve?
Here is one ingest consumer lag graph of the same time:
With events we can have discrete events that could include information from the backpressure system for more context. These can be overlaid on top of graphs:
Codecov Report
Attention: Patch coverage is 75.00000% with 1 lines in your changes are missing coverage. Please review.
Project coverage is 77.89%. Comparing base (
7eb565d) to head (51ae4cd). Report is 1 commits behind head on master.
Additional details and impacted files
@@ Coverage Diff @@
## master #71086 +/- ##
==========================================
- Coverage 77.89% 77.89% -0.01%
==========================================
Files 6525 6525
Lines 290597 290601 +4
Branches 50276 50276
==========================================
- Hits 226375 226372 -3
- Misses 57975 57979 +4
- Partials 6247 6250 +3
| Files | Coverage Δ | |
|---|---|---|
| src/sentry/processing/backpressure/health.py | 83.78% <75.00%> (-0.51%) |
:arrow_down: |