Add detailed debugging and verification steps to Kubernetes workflow

- Included checks for job existence and creation, along with status verification.
- Added endpoint availability tests and enhanced logging for database migration and build processes.
- Improved job wait process with additional status outputs and timestamps for better tracking.
This commit is contained in:
joereg4 2025-09-21 13:43:25 -05:00
parent d4471a1ca0
commit 93f667bfc2
1 changed files with 41 additions and 2 deletions

View File

@ -80,6 +80,10 @@ jobs:
- name: K8s Apply
run: |
make kustomize-local-apply
echo "=== Verifying job creation ==="
kubectl get jobs -n pokeapi
echo "=== Checking if load-graphql job exists ==="
kubectl get job load-graphql -n pokeapi || echo "Job not found - this is the problem!"
kubectl proxy &
sleep 1
bash Resources/scripts/wait.sh http://localhost:8001/api/v1/namespaces/pokeapi/services/pokeapi/proxy/api/v2/
@ -87,10 +91,23 @@ jobs:
run: |
kubectl config set-context --current --namespace pokeapi
kubectl describe deployment
echo "=== All resources in namespace ==="
kubectl get all -n pokeapi
echo "=== Checking for job creation errors ==="
kubectl get events -n pokeapi --sort-by='.lastTimestamp' | tail -20
- name: Migrate and build data
run: |
echo "=== Starting database migration at $(date) ==="
make k8s-migrate
echo "=== Starting database build at $(date) ==="
make k8s-build-db
echo "=== Database build completed at $(date) ==="
echo "=== Testing endpoint availability ==="
echo "Testing basic PokeAPI endpoint:"
curl -f -s --max-time 10 -o /dev/null -w "HTTP Status: %{http_code}\n" http://localhost:8001/api/v1/namespaces/pokeapi/services/pokeapi/proxy/api/v2/pokemon/1/ || echo "Basic endpoint failed"
echo "Testing pal-park-area endpoint:"
curl -f -s --max-time 10 -o /dev/null -w "HTTP Status: %{http_code}\n" http://localhost:8001/api/v1/namespaces/pokeapi/services/pokeapi/proxy/api/v2/pal-park-area/5/ || echo "Pal-park-area endpoint failed"
echo "=== Waiting for pal-park-area endpoint ==="
bash Resources/scripts/wait.sh http://localhost:8001/api/v1/namespaces/pokeapi/services/pokeapi/proxy/api/v2/pal-park-area/5/
- name: Debug pod status before GraphQL load
run: |
@ -102,16 +119,37 @@ jobs:
kubectl get jobs -n pokeapi
echo "=== PokeAPI Logs (last 10 lines) ==="
kubectl logs -l component=pokeapi -n pokeapi --tail=10 || echo "No PokeAPI logs yet"
echo "=== Checking if pal-park-area data exists ==="
kubectl exec -n pokeapi deployment/pokeapi -- python manage.py shell -c "from pokemon_v2.models import PalParkArea; print(f'PalParkArea count: {PalParkArea.objects.count()}')" || echo "Failed to check PalParkArea data"
- name: Debug network connectivity
run: |
echo "=== Testing PokeAPI connectivity ==="
kubectl run debug-pod --image=curlimages/curl --rm -i --restart=Never -- curl -f -s --max-time 10 http://pokeapi:80/api/v2/pokemon/1/ || echo "PokeAPI not reachable"
- name: Create job manually if missing
run: |
echo "=== Checking if job exists before manual creation ==="
if ! kubectl get job load-graphql -n pokeapi >/dev/null 2>&1; then
echo "Job missing - attempting manual creation"
kubectl apply -f Resources/k8s/kustomize/base/jobs/load-graphql.yaml
echo "=== Job created, checking status ==="
kubectl get job load-graphql -n pokeapi
else
echo "Job already exists"
fi
echo "=== Testing endpoint from within cluster ==="
kubectl run test-endpoint --image=curlimages/curl --rm -i --restart=Never -- curl -f -s --max-time 10 -o /dev/null -w "HTTP Status: %{http_code}\n" http://pokeapi:80/api/v2/pal-park-area/5/ || echo "Endpoint not accessible from within cluster"
- name: K8s wait for job
run: |
kubectl wait --timeout=1200s --for=condition=complete job/load-graphql || {
echo "=== JOB TIMEOUT DEBUG INFO ==="
echo "=== Starting job wait at $(date) ==="
echo "=== Job status before wait ==="
kubectl get job load-graphql -n pokeapi -o wide
echo "=== Pod status before wait ==="
kubectl get pods -n pokeapi -o wide
kubectl wait --timeout=600s --for=condition=complete job/load-graphql || {
echo "=== JOB TIMEOUT DEBUG INFO at $(date) ==="
echo "Job status:"
kubectl get job load-graphql -n pokeapi -o yaml
echo "Pod status:"
@ -122,6 +160,7 @@ jobs:
kubectl logs job/load-graphql -n pokeapi -c load-graphql --tail=20 || echo "Main container not started"
exit 1
}
echo "=== Job completed successfully at $(date) ==="
last_command=$(kubectl get job -o jsonpath='{.status.succeeded}' load-graphql)
test "$last_command" -eq 1
- name: Get GQL output