consolidate small prs

ansible.cfg

@@ -29,3 +29,7 @@ nocows = True
 # Additional ssh options for OpenShift Ansible
 [ssh_connection]
 pipelining = True
+# shorten the ControlPath which is often too long; when it is,
+# ssh connection reuse silently fails, making everything slower.
+control_path = %(directory)s/%%h-%%r
+

images/installer/Dockerfile.rhel7

@@ -2,16 +2,20 @@ FROM openshift3/playbook2image
 
 MAINTAINER OpenShift Team <dev@lists.openshift.redhat.com>
 
+# override env vars from base image
+ENV SUMMARY="OpenShift's installation and configuration tool" \
+    DESCRIPTION="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster"
+
 LABEL name="openshift3/ose-ansible" \
-      summary="OpenShift's installation and configuration tool" \
-      description="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster" \
+      summary="$SUMMARY" \
+      description="$DESCRIPTION" \
       url="https://github.com/openshift/openshift-ansible" \
       io.k8s.display-name="openshift-ansible" \
-      io.k8s.description="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster" \
+      io.k8s.description="$DESCRIPTION" \
       io.openshift.expose-services="" \
       io.openshift.tags="openshift,install,upgrade,ansible" \
       com.redhat.component="aos3-installation-docker" \
-      version="v3.4.1" \
+      version="v3.6.0" \
       release="1" \
       architecture="x86_64"
 
@@ -20,13 +24,18 @@ LABEL name="openshift3/ose-ansible" \
 # because all content and dependencies (like 'oc') is already
 # installed via yum.
 USER root
-RUN INSTALL_PKGS="atomic-openshift-utils atomic-openshift-clients python-boto" && \
+RUN INSTALL_PKGS="atomic-openshift-utils atomic-openshift-clients python-boto skopeo openssl java-1.8.0-openjdk-headless httpd-tools" && \
     yum repolist > /dev/null && \
-    yum-config-manager --enable rhel-7-server-ose-3.4-rpms && \
+    yum-config-manager --enable rhel-7-server-ose-3.6-rpms && \
     yum-config-manager --enable rhel-7-server-rh-common-rpms && \
     yum install -y $INSTALL_PKGS && \
     yum clean all
 
+# The symlinks below are a (hopefully temporary) hack to work around the fact that this
+# image is based on python s2i which uses the python27 SCL instead of system python,
+# and so the system python modules we need would otherwise not be in the path.
+RUN ln -s /usr/lib/python2.7/site-packages/{boto,passlib} /opt/app-root/lib64/python2.7/
+
 USER ${USER_UID}
 
 # The playbook to be run is specified via the PLAYBOOK_FILE env var.
@@ -36,6 +45,7 @@ USER ${USER_UID}
 # $APP_HOME by the 'assemble' script, we set the WORK_DIR env var to the
 # location of openshift-ansible.
 ENV PLAYBOOK_FILE=playbooks/byo/openshift_facts.yml \
+    ANSIBLE_CONFIG=/usr/share/atomic-openshift-utils/ansible.cfg \
     WORK_DIR=/usr/share/ansible/openshift-ansible \
     OPTS="-v"
 

roles/openshift_health_checker/callback_plugins/zz_failure_summary.py

@@ -39,7 +39,8 @@ def v2_playbook_on_start(self, playbook):
 
     def v2_runner_on_failed(self, result, ignore_errors=False):
         super(CallbackModule, self).v2_runner_on_failed(result, ignore_errors)
-        self.__failures.append(dict(result=result, ignore_errors=ignore_errors))
+        if not ignore_errors:
+            self.__failures.append(dict(result=result, ignore_errors=ignore_errors))
 
     def v2_playbook_on_stats(self, stats):
         super(CallbackModule, self).v2_playbook_on_stats(stats)

roles/openshift_health_checker/openshift_checks/docker_storage.py

@@ -17,7 +17,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
     tags = ["pre-install", "health", "preflight"]
 
     dependencies = ["python-docker-py"]
-    storage_drivers = ["devicemapper", "overlay2"]
+    storage_drivers = ["devicemapper", "overlay", "overlay2"]
     max_thinpool_data_usage_percent = 90.0
     max_thinpool_meta_usage_percent = 90.0
 

roles/openshift_health_checker/openshift_checks/logging/kibana.py

@@ -62,7 +62,7 @@ def _verify_url_internal(self, url, task_vars):
             # TODO(lmeyer): give users option to validate certs
             status_code=302,
         )
-        result = self.execute_module('uri', args, task_vars)
+        result = self.execute_module('uri', args, None, task_vars)
         if result.get('failed'):
             return result['msg']
         return None

roles/openshift_health_checker/openshift_checks/logging/logging.py

@@ -54,12 +54,12 @@ def not_running_pods(pods):
         """Returns: list of pods not in a ready and running state"""
         return [
             pod for pod in pods
-            if any(
+            if not pod.get("status", {}).get("containerStatuses") or any(
                 container['ready'] is False
                 for container in pod['status']['containerStatuses']
             ) or not any(
                 condition['type'] == 'Ready' and condition['status'] == 'True'
-                for condition in pod['status']['conditions']
+                for condition in pod['status'].get('conditions', [])
             )
         ]
 
@@ -78,7 +78,7 @@ def exec_oc(execute_module=None, namespace="logging", cmd_str="", extra_args=Non
             "extra_args": list(extra_args) if extra_args else [],
         }
 
-        result = execute_module("ocutil", args, task_vars)
+        result = execute_module("ocutil", args, None, task_vars)
         if result.get("failed"):
             msg = (
                 'Unexpected error using `oc` to validate the logging stack components.\n'

roles/openshift_health_checker/test/docker_storage_test.py

@@ -65,8 +65,8 @@ def test_is_active(is_containerized, group_names, is_active):
         dict(info={
             "Driver": "overlay",
         }),
-        True,
-        ["unsupported Docker storage driver"],
+        False,
+        [],
     ),
     (
         dict(info={

roles/openshift_health_checker/test/kibana_test.py

@@ -169,7 +169,7 @@ def test_get_kibana_url(route, expect_url, expect_error):
     ),
 ])
 def test_verify_url_internal_failure(exec_result, expect):
-    check = Kibana(execute_module=lambda module_name, args, task_vars: dict(failed=True, msg=exec_result))
+    check = Kibana(execute_module=lambda module_name, args, tmp, task_vars: dict(failed=True, msg=exec_result))
     check._get_kibana_url = lambda task_vars: ('url', None)
 
     error = check._check_kibana_route({})

roles/openshift_health_checker/test/logging_check_test.py

@@ -50,6 +50,16 @@ def assert_error(error, expect_error):
     }
 }
 
+plain_kibana_pod_no_containerstatus = {
+    "metadata": {
+        "labels": {"component": "kibana", "deploymentconfig": "logging-kibana"},
+        "name": "logging-kibana-1",
+    },
+    "status": {
+        "conditions": [{"status": "True", "type": "Ready"}],
+    }
+}
+
 fluentd_pod_node1 = {
     "metadata": {
         "labels": {"component": "fluentd", "deploymentconfig": "logging-fluentd"},
@@ -80,7 +90,7 @@ def assert_error(error, expect_error):
     ("Permission denied", "Unexpected error using `oc`"),
 ])
 def test_oc_failure(problem, expect):
-    def execute_module(module_name, args, task_vars):
+    def execute_module(module_name, args, tmp, task_vars):
         if module_name == "ocutil":
             return dict(failed=True, result=problem)
         return dict(changed=False)
@@ -135,3 +145,23 @@ def test_get_pods_for_component(pod_output, expect_pods, expect_error):
         {}
     )
     assert_error(error, expect_error)
+
+
+@pytest.mark.parametrize('name, pods, expected_pods', [
+    (
+        'test single pod found, scheduled, but no containerStatuses field',
+        [plain_kibana_pod_no_containerstatus],
+        [plain_kibana_pod_no_containerstatus],
+    ),
+    (
+        'set of pods has at least one pod with containerStatuses (scheduled); should still fail',
+        [plain_kibana_pod_no_containerstatus, plain_kibana_pod],
+        [plain_kibana_pod_no_containerstatus],
+    ),
+
+], ids=lambda argvals: argvals[0])
+def test_get_not_running_pods_no_container_status(name, pods, expected_pods):
+    check = canned_loggingcheck(lambda exec_module, namespace, cmd, args, task_vars: '')
+    result = check.not_running_pods(pods)
+
+    assert result == expected_pods

utils/etc/ansible.cfg

@@ -28,3 +28,9 @@ deprecation_warnings = False
 # remote_tmp - set if provided by user (cli)
 # ssh_args - set if provided by user (cli)
 # control_path
+
+# Additional ssh options for OpenShift Ansible
+[ssh_connection]
+# shorten the ControlPath which is often too long; when it is,
+# ssh connection reuse silently fails, making everything slower.
+control_path = %(directory)s/%%h-%%r