Add healthchecks (#383)

* added additional configuration variable for director EXTRA_HOSTS_SUFFIX
if there are extra hosts defined for the director it will pass them to the services

* added generic python based curl for docker

* added healthcheck for director, webserver, modeling webserver, jupyter notebook and paraviewweb

* some notes for hybrid swarm

* pylint

Author: sanderegg

scripts/docker/healthcheck_curl_host.py

@@ -0,0 +1,9 @@
+import os
+import sys
+
+try:
+    from urllib.request import urlopen
+except ImportError:
+    from urllib2 import urlopen
+
+print(0 if urlopen("{host}{baseurl}".format(host=sys.argv[1], baseurl=os.environ.get("SIMCORE_NODE_BASEPATH", "")) ).getcode() == 200 else 1)

services/director/Dockerfile

@@ -22,6 +22,11 @@ ENV REGISTRY_URL = ''
 ENV PUBLISHED_HOST_NAME=''
 ENV EXTRA_HOSTS_SUFFIX = 'undefined'
 
+# ----------------------------------------------------------------
+COPY --chown=scu:scu scripts/docker/healthcheck_curl_host.py /healthcheck/healthcheck_curl_host.py
+# will start to run after interval every interval. fails after timeout. fail do not count if during start-period. will do # retries
+HEALTHCHECK --interval=10s --timeout=30s --start-period=1s --retries=3 CMD [ "python3", "/healthcheck/healthcheck_curl_host.py", "http://localhost:8001/v0/" ]
+
 WORKDIR /home/scu
 
 # -------------------------- Build stage -------------------

services/dy-3dvis/simcoreparaviewweb/Dockerfile

@@ -40,6 +40,10 @@ RUN curl -L https://raw.githubusercontent.com/yyuu/pyenv-installer/master/bin/py
 COPY --from=visualizer /home/node/visualizer/dist /opt/paraview/install/share/paraview-5.5/web/visualizer/www
 WORKDIR /home/root
 # ----------------------------------------------------------------
+COPY scripts/docker/healthcheck_curl_host.py /healthcheck/healthcheck_curl_host.py
+# will start to run after interval every interval. fails after timeout. fail do not count if during start-period. will do # retries
+HEALTHCHECK --interval=10s --timeout=30s --start-period=1s --retries=3 CMD [ "python3", "/healthcheck/healthcheck_curl_host.py", "http://localhost:80/visualizer" ]
+
 # set up oSparc env variables
 ENV SIMCORE_NODE_UUID="-1" \
     SIMCORE_USER_ID="-1" \

services/dy-jupyter/Dockerfile

@@ -31,6 +31,9 @@ ENV SIMCORE_NODE_UUID="-1" \
     POSTGRES_PASSWORD="-1" \
     POSTGRES_DB="-1"
 # ----------------------------------------------------------------
+COPY --chown=jovyan:users scripts/docker/healthcheck_curl_host.py /healthcheck/healthcheck_curl_host.py
+# will start to run after interval every interval. fails after timeout. fail do not count if during start-period. will do # retries
+HEALTHCHECK --interval=10s --timeout=30s --start-period=1s --retries=3 CMD [ "python3", "/healthcheck/healthcheck_curl_host.py", "http://localhost:8888" ]
 # configure notebook
 RUN pip install --upgrade pip && \
     pip install jupyter_contrib_nbextensions && \

services/dy-modeling/README.md

@@ -1 +1,6 @@
-# Dynamic modeling service using a [qooxdoo](http://qooxdoo.org/) + [three.js](https://threejs.org/) client and a [node.js] (https://nodejs.org/) server
+# Dynamic modeling service using a [qooxdoo](http://qooxdoo.org/) + [three.js](https://threejs.org/) client and a [node.js] (https://nodejs.org/) server
+
+## Hybrid swarm (Windows tips)
+### multiple network adapters
+Check the Hyper-V manager in the switch manager. When joining the swarm docker creates a virtual network named as the network ID. Check that this
+network actually uses the right adapter.

services/dy-modeling/server/Dockerfile

@@ -16,7 +16,10 @@ ARG s4l_port_mod="9096"
 ENV CS_S4L_HOSTNAME $s4l_ip
 ENV CS_S4L_PORT_APP $s4l_port_app
 ENV CS_S4L_PORT_MOD $s4l_port_mod
-
+# ----------------------------------------------------------------
+COPY scripts/docker/healthcheck_curl_host.py /healthcheck/healthcheck_curl_host.py
+# will start to run after interval every interval. fails after timeout. fail do not count if during start-period. will do # retries
+HEALTHCHECK --interval=10s --timeout=30s --start-period=1s --retries=3 CMD [ "python", "/healthcheck/healthcheck_curl_host.py", "http://localhost:9090" ]
 # + /home/node/          json files
 #    + source-output/    client source files
 #    + source/  server   source files

services/web/Dockerfile

@@ -23,6 +23,11 @@ ENV IS_CONTAINER_CONTEXT Yes
 
 EXPOSE 8080
 
+# ----------------------------------------------------------------
+COPY --chown=scu:scu scripts/docker/healthcheck_curl_host.py /healthcheck/healthcheck_curl_host.py
+# will start to run after interval every interval. fails after timeout. fail do not count if during start-period. will do # retries
+HEALTHCHECK --interval=10s --timeout=30s --start-period=1s --retries=3 CMD [ "python", "/healthcheck/healthcheck_curl_host.py", "http://localhost:8080" ]
+
 # -------------------------- Build stage -------------------
 #
 # + /home/scu/              $HOME

services/web/server/src/simcore_service_webserver/director/handlers.py

@@ -34,11 +34,7 @@ def _resolve_url(request: web.Request) -> URL:
 
 @login_required
 async def services_get(request: web.Request) -> web.Response:
-    params, query, body = await extract_and_validate(request)
-
-    assert not params
-    assert query
-    assert not body
+    _, _, _ = await extract_and_validate(request)
 
     url = _resolve_url(request)