♻️🎨Dockerfile: fix start period/start interval healthchecks (#7557)

sanderegg · web-flow · commit e6f13b112666 · 2025-04-22T16:38:05.000+02:00
diff --git a/scripts/docker/healthcheck_curl_host.py b/scripts/docker/healthcheck_curl_host.py
@@ -6,7 +6,8 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
                 CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
diff --git a/services/agent/Dockerfile b/services/agent/Dockerfile
@@ -148,10 +148,13 @@ COPY --chown=scu:scu services/agent/docker services/agent/docker
 RUN chmod +x services/agent/docker/*.sh
 
 
-HEALTHCHECK --interval=30s \
-  --timeout=20s \
-  --start-period=30s \
-  --retries=3 \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
+  --timeout=5s \
+  --start-period=20s \
+  --start-interval=1s \
+  --retries=5 \
   CMD ["python3", "services/agent/docker/healthcheck.py", "http://localhost:8000/health"]
 
 EXPOSE 8000
diff --git a/services/agent/docker/healthcheck.py b/services/agent/docker/healthcheck.py
@@ -6,7 +6,8 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
                 CMD python3 docker/healthcheck.py http://localhost:8000/
 ```
@@ -15,6 +16,7 @@
     1. why not to use curl instead of a python script?
         - SEE https://blog.sixeyed.com/docker-healthchecks-why-not-to-use-curl-or-iwr/
 """
+
 import os
 import sys
 from urllib.request import urlopen
diff --git a/services/api-server/Dockerfile b/services/api-server/Dockerfile
@@ -143,7 +143,8 @@ RUN chmod +x services/api-server/docker/*.sh
 
 HEALTHCHECK --interval=10s \
   --timeout=5s \
-  --start-period=5s \
+  --start-period=20s \
+  --start-interval=1s \
   --retries=5 \
   CMD ["python3", "services/api-server/docker/healthcheck.py", "http://localhost:8000/"]
 
diff --git a/services/api-server/docker/healthcheck.py b/services/api-server/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/autoscaling/Dockerfile b/services/autoscaling/Dockerfile
@@ -156,9 +156,12 @@ COPY --chown=scu:scu services/autoscaling/docker services/autoscaling/docker
 RUN chmod +x services/autoscaling/docker/*.sh
 
 
-HEALTHCHECK --interval=10s \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
   --timeout=5s \
-  --start-period=5s \
+  --start-period=20s \
+  --start-interval=1s \
   --retries=5 \
   CMD ["python3", "services/autoscaling/docker/healthcheck.py", "http://localhost:8000/"]
 
diff --git a/services/autoscaling/docker/healthcheck.py b/services/autoscaling/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/catalog/Dockerfile b/services/catalog/Dockerfile
@@ -140,9 +140,12 @@ COPY --chown=scu:scu --from=prod-only-deps  ${VIRTUAL_ENV} ${VIRTUAL_ENV}
 COPY --chown=scu:scu services/catalog/docker services/catalog/docker
 RUN chmod +x services/catalog/docker/*.sh
 
-HEALTHCHECK --interval=10s \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
   --timeout=5s \
-  --start-period=5s \
+  --start-period=20s \
+  --start-interval=1s \
   --retries=5 \
   CMD ["python3", "services/catalog/docker/healthcheck.py", "http://localhost:8000/"]
 
diff --git a/services/catalog/docker/healthcheck.py b/services/catalog/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/clusters-keeper/Dockerfile b/services/clusters-keeper/Dockerfile
@@ -159,9 +159,12 @@ COPY --chown=scu:scu services/clusters-keeper/docker services/clusters-keeper/do
 RUN chmod +x services/clusters-keeper/docker/*.sh
 
 
-HEALTHCHECK --interval=10s \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
   --timeout=5s \
-  --start-period=5s \
+  --start-period=20s \
+  --start-interval=1s \
   --retries=5 \
   CMD ["python3", "services/clusters-keeper/docker/healthcheck.py", "http://localhost:8000/"]
 
diff --git a/services/clusters-keeper/docker/healthcheck.py b/services/clusters-keeper/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/dask-sidecar/Dockerfile b/services/dask-sidecar/Dockerfile
@@ -151,10 +151,12 @@ RUN chmod +x services/dask-sidecar/docker/*.sh
 # In order to have the same healty entrypoint port
 # make sure dask worker is started as ``dask-worker --dashboard-address 8787``.
 # Otherwise the worker will take random ports to serve the /health entrypoint.
+# https://docs.docker.com/reference/dockerfile/#healthcheck
 HEALTHCHECK \
   --interval=10s \
   --timeout=5s \
-  --start-period=5s \
+  --start-period=20s \
+  --start-interval=1s \
   --retries=5 \
   CMD ["curl", "-Lf", "http://127.0.0.1:8787/health"]
 
diff --git a/services/datcore-adapter/Dockerfile b/services/datcore-adapter/Dockerfile
@@ -141,9 +141,12 @@ COPY --chown=scu:scu services/datcore-adapter/docker services/datcore-adapter/do
 RUN chmod +x services/datcore-adapter/docker/*.sh
 
 
-HEALTHCHECK --interval=10s \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
   --timeout=5s \
-  --start-period=5s \
+  --start-period=20s \
+  --start-interval=1s \
   --retries=5 \
   CMD ["python3", "services/datcore-adapter/docker/healthcheck.py", "http://localhost:8000/v0/live"]
 
diff --git a/services/datcore-adapter/docker/healthcheck.py b/services/datcore-adapter/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/director-v2/Dockerfile b/services/director-v2/Dockerfile
@@ -141,10 +141,13 @@ COPY --chown=scu:scu services/director-v2/docker services/director-v2/docker
 RUN chmod +x services/director-v2/docker/*.sh
 
 
-HEALTHCHECK --interval=30s \
-  --timeout=20s \
-  --start-period=30s \
-  --retries=3 \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
+  --timeout=5s \
+  --start-period=20s \
+  --start-interval=1s \
+  --retries=5 \
   CMD ["python3", "services/director-v2/docker/healthcheck.py", "http://localhost:8000/"]
 
 ENTRYPOINT [ "/bin/sh", "services/director-v2/docker/entrypoint.sh" ]
diff --git a/services/director-v2/docker/healthcheck.py b/services/director-v2/docker/healthcheck.py
@@ -6,15 +6,17 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
     1. why not to use curl instead of a python script?
         - SEE https://blog.sixeyed.com/docker-healthchecks-why-not-to-use-curl-or-iwr/
 """
+
 import os
 import sys
 from urllib.request import urlopen
diff --git a/services/director/Dockerfile b/services/director/Dockerfile
@@ -137,9 +137,11 @@ COPY --chown=scu:scu services/director/docker services/director/docker
 RUN chmod +x services/director/docker/*.sh
 
 
-HEALTHCHECK --interval=10s \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
   --timeout=5s \
-  --start-period=30s \
+  --start-period=20s \
   --start-interval=1s \
   --retries=5 \
   CMD ["python3", "/home/scu/services/director/docker/healthcheck.py", "http://localhost:8000/v0/"]
diff --git a/services/director/docker/healthcheck.py b/services/director/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/docker-api-proxy/Dockerfile b/services/docker-api-proxy/Dockerfile
@@ -22,10 +22,11 @@ RUN apk add --no-cache \
   su-exec
 
 # Health check to ensure the proxy is running
+# https://docs.docker.com/reference/dockerfile/#healthcheck
 HEALTHCHECK \
   --interval=10s \
   --timeout=5s \
-  --start-period=30s \
+  --start-period=20s \
   --start-interval=1s \
   --retries=5 \
   CMD curl http://localhost:8888/version || exit 1
diff --git a/services/dynamic-scheduler/Dockerfile b/services/dynamic-scheduler/Dockerfile
@@ -141,10 +141,13 @@ COPY --chown=scu:scu services/dynamic-scheduler/docker services/dynamic-schedule
 RUN chmod +x services/dynamic-scheduler/docker/*.sh
 
 
-HEALTHCHECK --interval=30s \
-  --timeout=20s \
-  --start-period=30s \
-  --retries=3 \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
+  --timeout=5s \
+  --start-period=20s \
+  --start-interval=1s \
+  --retries=5 \
   CMD ["python3", "services/dynamic-scheduler/docker/healthcheck.py", "http://localhost:8000/health"]
 
 ENTRYPOINT [ "/bin/sh", "services/dynamic-scheduler/docker/entrypoint.sh" ]
diff --git a/services/dynamic-scheduler/docker/healthcheck.py b/services/dynamic-scheduler/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/dynamic-sidecar/Dockerfile b/services/dynamic-sidecar/Dockerfile
@@ -196,6 +196,10 @@ COPY --chown=scu:scu --from=prod-only-deps  ${DYNAMIC_SIDECAR_DY_VOLUMES_MOUNT_D
 COPY --chown=scu:scu services/dynamic-sidecar/docker services/dynamic-sidecar/docker
 RUN chmod +x services/dynamic-sidecar/docker/*.sh
 
+# NOTE: the start period of 3 minutes is to allow the dynamic-sidecar
+# enough time to connect to the externald dependencies. some times the docker
+# networks take time to get created
+# https://docs.docker.com/reference/dockerfile/#healthcheck
 HEALTHCHECK \
   --interval=10s \
   --timeout=5s \
diff --git a/services/efs-guardian/Dockerfile b/services/efs-guardian/Dockerfile
@@ -175,9 +175,11 @@ COPY services/efs-guardian/docker services/efs-guardian/docker
 RUN chmod +x services/efs-guardian/docker/*.sh
 
 
-HEALTHCHECK --interval=10s \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
   --timeout=5s \
-  --start-period=30s \
+  --start-period=20s \
   --start-interval=1s \
   --retries=5 \
   CMD ["python3", "services/efs-guardian/docker/healthcheck.py", "http://localhost:8000/"]
diff --git a/services/efs-guardian/docker/healthcheck.py b/services/efs-guardian/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/invitations/Dockerfile b/services/invitations/Dockerfile
@@ -141,10 +141,13 @@ COPY --chown=scu:scu services/invitations/docker services/invitations/docker
 RUN chmod +x services/invitations/docker/*.sh
 
 
-HEALTHCHECK --interval=30s \
-  --timeout=20s \
-  --start-period=30s \
-  --retries=3 \
+# https://docs.docker.com/reference/dockerfile/#healthcheck
+HEALTHCHECK \
+  --interval=10s \
+  --timeout=5s \
+  --start-period=20s \
+  --start-interval=1s \
+  --retries=5 \
   CMD ["python3", "services/invitations/docker/healthcheck.py", "http://localhost:8000/"]
 
 ENTRYPOINT [ "/bin/sh", "services/invitations/docker/entrypoint.sh" ]
diff --git a/services/invitations/docker/healthcheck.py b/services/invitations/docker/healthcheck.py
@@ -6,9 +6,10 @@
     COPY --chown=scu:scu docker/healthcheck.py docker/healthcheck.py
     HEALTHCHECK --interval=30s \
                 --timeout=30s \
-                --start-period=1s \
+                --start-period=20s \
+                --start-interval=1s \
                 --retries=3 \
-                CMD python3 docker/healthcheck.py http://localhost:8000/
+                CMD python3 docker/healthcheck.py http://localhost:8080/v0/
 ```
 
 Q&A:
diff --git a/services/migration/Dockerfile b/services/migration/Dockerfile
@@ -95,11 +95,13 @@ RUN chown -R scu /home/scu
 COPY --chown=scu:scu --from=build ${VIRTUAL_ENV} ${VIRTUAL_ENV}
 COPY --chown=scu:scu services/migration/docker services/migration/docker
 
+# https://docs.docker.com/reference/dockerfile/#healthcheck
 HEALTHCHECK \
-  --interval=5s \
-  --timeout=10s \
-  --start-period=1ms \
-  --retries=50 \
+  --interval=10s \
+  --timeout=5s \
+  --start-period=20s \
+  --start-interval=1s \
+  --retries=5 \
   CMD [ "/bin/sh", "/home/scu/services/migration/docker/healthcheck.sh"]
 
 ENTRYPOINT [ "/bin/sh", "services/migration/docker/entrypoint.sh" ]
diff --git a/services/payments/Dockerfile b/services/payments/Dockerfile
diff --git a/services/payments/docker/healthcheck.py b/services/payments/docker/healthcheck.py
diff --git a/services/resource-usage-tracker/Dockerfile b/services/resource-usage-tracker/Dockerfile
diff --git a/services/resource-usage-tracker/docker/healthcheck.py b/services/resource-usage-tracker/docker/healthcheck.py
diff --git a/services/storage/Dockerfile b/services/storage/Dockerfile
diff --git a/services/web/Dockerfile b/services/web/Dockerfile
