feat: Add health check endpoints for deployment verification #948
Conversation
cfsmp3
requested review from
canihavesomecoffee and
thealphadollar
as code owners
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #948 +/- ##
==========================================
+ Coverage 86.88% 87.02% +0.14%
==========================================
Files 35 36 +1
Lines 3759 3800 +41
Branches 767 774 +7
==========================================
+ Hits 3266 3307 +41
Misses 355 355
Partials 138 138
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
mod_health/controllers.py
Outdated
| db.remove() | ||
| return {'status': 'ok'} | ||
| except Exception as e: | ||
| return {'status': 'error', 'message': str(e)} |
There was a problem hiding this comment.
We don't want to just dump the exception message here, it might contain sensitive information. Better log the exception and return a generic error message.
There was a problem hiding this comment.
Good catch! Fixed - now logging the exception server-side with g.log.exception() and returning a generic "Database connection failed" message to the client.
|
Also, you might want to rebase the branch :) |
Add /health, /health/live, and /health/ready endpoints to enable deployment verification and monitoring: - /health: Returns 200 if database and config checks pass, 503 otherwise - /health/live: Always returns 200 if Flask is responding (liveness probe) - /health/ready: Same as /health (readiness probe) These endpoints are essential for implementing safe deployments with automatic rollback - the deployment pipeline can verify the app is healthy after deploying new code. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Log database exceptions server-side with g.log.exception() - Return generic "Database connection failed" message to client - Add comment explaining db.remove() connection pool cleanup - Update test to expect generic error message 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
cfsmp3
force-pushed
the
feat/health-endpoint
branch
from
ab0cf74 to
7f26810
Compare
|
Rebased on master and addressed review comments in the latest commit. CI is now running. |
g.log is only available during HTTP requests (set in before_request), but current_app.logger works in any app context including tests. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
|
Summary
Adds health check endpoints to enable deployment verification and monitoring:
/health- Returns 200 if all critical checks pass (database, config), 503 otherwise/health/live- Liveness probe, always returns 200 if Flask is responding/health/ready- Readiness probe, same as/healthWhy This Is Needed
Currently, the deployment pipeline has no way to verify that a deployment was successful. If code is deployed that crashes on startup or can't connect to the database, the platform becomes "bricked" with no automatic recovery.
These endpoints enable:
/healthto confirm the app started successfullyTest Plan
/healthreturns 200 when database and config are OK/healthreturns 503 when database fails/healthreturns 503 when config is missing/health/livealways returns 200/health/readybehaves same as/healthExample Response
{ "status": "healthy", "timestamp": "2025-12-23T15:30:00Z", "checks": { "database": {"status": "ok"}, "config": {"status": "ok"} } }Related
This is part of a larger effort to make deployments more reliable. Future PRs will:
🤖 Generated with Claude Code