Check: health checks | ComputeSphere Learn

Goal: Confirm you can say what a health check protects, write one that isn't fragile, and fix a restart loop.

Four questions: three on Module 4, one on Module 2. Every answer explains itself, right or wrong.

Your /healthz checks the database on every call. The database is unreachable for three minutes. What happens on ComputeSphere?Nothing: the check knows it's the database, not the appEvery spherelet leaves traffic, then gets restartedThe service is deleted and redeployed

Where should the app report that its database is unreachable, without failing the health check?On a separate /status endpoint and in its logsIn /healthz, with a 503Nowhere: it isn't the app's problem

An update stays on Redeploying. The deploy log repeats Unhealthy, and the runtime log shows the app starting over every two minutes, never reaching its ready line. It takes three minutes to start. Initial delay is 10 and Check interval 10. What do you do?Roll back and look for a bugRaise Failure threshold to 20Raise Initial delay above three minutes, then redeploy

From Module 2: a new service fails, and the deploy log says Restarting repeatedly. Where do you look next?The runtime log, for the last lines before it exitsThe build logThe Metrics tab

Retry as often as you like.