Create Detector State Model

[saponifi3d]

Description

Add the detector state model.

This model is used to track the state of the detector for Alerts Create Issues

[github-actions]

🚨 Warning: This pull request contains Frontend and Backend changes!

It's discouraged to make changes to Sentry's Frontend and Backend in a single pull request. The Frontend and Backend are not atomically deployed. If the changes are interdependent of each other, they must be separated into two pull requests and be made forward or backwards compatible, such that the Backend or Frontend can be safely deployed independently.

Have questions? Please ask in the #discuss-dev-infra channel.

[github-actions]

This PR has a migration; here is the generated SQL for src/sentry/workflow_engine/migrations/0007_detector_state.py ()

--
-- Create model DetectorState
--
CREATE TABLE "workflow_engine_detectorstate" ("id" bigint NOT NULL PRIMARY KEY GENERATED BY DEFAULT AS IDENTITY, "date_updated" timestamp with time zone NOT NULL, "date_added" timestamp with time zone NOT NULL, "detector_group_key" varchar(200) NULL, "status" varchar(200) NOT NULL, "state" varchar(200) NULL, "detector_id" bigint NOT NULL);
ALTER TABLE "workflow_engine_detectorstate" ADD CONSTRAINT "workflow_engine_dete_detector_id_72d03dd7_fk_workflow_" FOREIGN KEY ("detector_id") REFERENCES "workflow_engine_detector" ("id") DEFERRABLE INITIALLY DEFERRED NOT VALID;
ALTER TABLE "workflow_engine_detectorstate" VALIDATE CONSTRAINT "workflow_engine_dete_detector_id_72d03dd7_fk_workflow_";
CREATE INDEX CONCURRENTLY "workflow_engine_detectorstate_detector_id_72d03dd7" ON "workflow_engine_detectorstate" ("detector_id");

[JoshFerge]

nit: StatusType? assuming there will be some kind of Type var for state too

[wedamija]

Is the idea here that for a metric alert, we'd have

status: active, state: warning
status: active, state: critical
status: inactive, state: null?

or status:ok/warning/critical?

[saponifi3d]

The former, or close to it. i was thinking:

status: active, state: warning
status: active, state: critical
status: inactive, state: ok

---

I was trying to design for a bunch of cases tho.. for performance issues, i was thinking:

status:inactive, state: ok
status:active, state: detected_n+1

Since those have an autoresolve, we could flip the status to inactive and leave the state:

status: inactive, state: detected_n+1

that could give us some other insights, and if we keep seeing the issue not be addressed we could add an escalation policy.

An escalation policy could be something like...

if status inactive -> active with !ok state
    update redis counter by detector / id.

if redis counter > escalation number
    trigger escalation

---

it also allows us to easily evaluate any kind of detector by the status without having to check it's state - but if we want to simplify for now, we could probably create a shared DetectorState.OK enum value. i've just found that a little harder to maintain across multiple teams / products than a required field (🤔 should probably just make it an active boolean though).

[saponifi3d]

^ pushed a quick update to it that i talked myself into in that comment 🤣 also addresses josh's concern about the status Type name.

[github-actions]

This PR has a migration; here is the generated SQL for src/sentry/workflow_engine/migrations/0008_detector_state.py ()

--
-- Create model DetectorState
--
CREATE TABLE "workflow_engine_detectorstate" ("id" bigint NOT NULL PRIMARY KEY GENERATED BY DEFAULT AS IDENTITY, "date_updated" timestamp with time zone NOT NULL, "date_added" timestamp with time zone NOT NULL, "detector_group_key" varchar(200) NULL, "active" boolean NOT NULL, "state" varchar(200) NOT NULL, "detector_id" bigint NOT NULL);
ALTER TABLE "workflow_engine_detectorstate" ADD CONSTRAINT "workflow_engine_dete_detector_id_72d03dd7_fk_workflow_" FOREIGN KEY ("detector_id") REFERENCES "workflow_engine_detector" ("id") DEFERRABLE INITIALLY DEFERRED NOT VALID;
ALTER TABLE "workflow_engine_detectorstate" VALIDATE CONSTRAINT "workflow_engine_dete_detector_id_72d03dd7_fk_workflow_";
CREATE INDEX CONCURRENTLY "workflow_engine_detectorstate_detector_id_72d03dd7" ON "workflow_engine_detectorstate" ("detector_id");

[codecov]

Codecov Report

Attention: Patch coverage is 94.11765% with 2 lines in your changes missing coverage. Please review.

✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
src/sentry/testutils/factories.py	66.66%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #78244   +/-   ##
=======================================
  Coverage   78.08%   78.08%           
=======================================
  Files        7067     7069    +2     
  Lines      311797   311831   +34     
  Branches    50958    50963    +5     
=======================================
+ Hits       243464   243491   +27     
- Misses      61988    61993    +5     
- Partials     6345     6347    +2