diff --git a/.github/workflows/prod_env_tests.yml b/.github/workflows/prod_env_tests.yml index 6b6b11278107c120c15e24a0760e42be173d4080..de1735eb2ec228f53bd4881c5cf5bac93f68fdbb 100644 --- a/.github/workflows/prod_env_tests.yml +++ b/.github/workflows/prod_env_tests.yml @@ -318,9 +318,13 @@ jobs: with: run: kubectl logs -l nodeType=localnode -c node --tail -1 + - name: Wait for CONTROLLER container to become ready + uses: jakejarvis/wait-action@master + with: + time: '60s' + - name: Run Healthcheck run: curl "http://172.17.0.1:5000/healthcheck" - continue-on-error: true # The first healthcheck could fail due to timeouts involved. When it times out, successful or not, everything should be running. - name: Update NodeLandscapeAggregator run: curl -X POST "http://172.17.0.1:5000/nla" diff --git a/kubernetes/templates/mipengine-controller.yaml b/kubernetes/templates/mipengine-controller.yaml index 582f3fdafbb36b6d557f585bb2ab7fb75c4b085a..5925000902bfd9f09f83b690a2cc49270ba375ce 100644 --- a/kubernetes/templates/mipengine-controller.yaml +++ b/kubernetes/templates/mipengine-controller.yaml @@ -67,14 +67,13 @@ spec: httpGet: path: /healthcheck port: 5000 - initialDelaySeconds: 30 + initialDelaySeconds: 60 periodSeconds: 30 timeoutSeconds: {{ .Values.controller.celery_tasks_timeout }} livenessProbe: httpGet: path: /healthcheck port: 5000 - initialDelaySeconds: 30 periodSeconds: 30 timeoutSeconds: {{ .Values.controller.celery_tasks_timeout }}