Platform-wide health check — all registered components

All registered platform components reachable on their health endpoints

PASS for all running services. FAIL with ECONNREFUSED = container not running. FAIL with HTTP error = container up but service unhealthy.

# Run from inside ops-tracker or any container on prysm_infra network
# Checks all components in ops_components that have a health_endpoint
# Usage: docker exec ops-tracker sh /tmp/this_script.sh

COMPONENTS=$(curl -s http://ops-tracker:3563/api/components)
echo $COMPONENTS | python3 -c "
import sys,json
data=json.load(sys.stdin)['data']
for c in data:
    ep=c.get('health_endpoint','')
    if ep and ep not in ('---',''):
        result='PASS' if __import__('subprocess').run(['wget','-q','-O','/dev/null','--timeout=5','--tries=1',ep],capture_output=True).returncode==0 else 'FAIL'
        print(result, c['name'], ep)
"