openshift_checks: refactor to internalize task_vars

Move task_vars into instance variable so we don't have to pass it
around everywhere. Also store tmp. Make sure both are filled in on
execute_module.

In the process, is_active became an instance method, and task_vars is
basically never used directly outside of test code.

19 files changed, 282 insertions, 293 deletions

diff --git a/roles/openshift_health_checker/openshift_checks/__init__.py b/roles/openshift_health_checker/openshift_checks/__init__.py
index c2cfd0175..40a28cde5 100644
--- a/roles/openshift_health_checker/openshift_checks/__init__.py
+++ b/roles/openshift_health_checker/openshift_checks/__init__.py
@@ -19,14 +19,21 @@ class OpenShiftCheckException(Exception):
 
 @six.add_metaclass(ABCMeta)
 class OpenShiftCheck(object):
-    """A base class for defining checks for an OpenShift cluster environment."""
+    """
+    A base class for defining checks for an OpenShift cluster environment.
+
+    Expect optional params: method execute_module, dict task_vars, and string tmp.
+    execute_module is expected to have a signature compatible with _execute_module
+    from ansible plugins/action/__init__.py, e.g.:
+    def execute_module(module_name=None, module_args=None, tmp=None, task_vars=None, *args):
+    This is stored so that it can be invoked in subclasses via check.execute_module("name", args)
+    which provides the check's stored task_vars and tmp.
+    """
 
-    def __init__(self, execute_module=None):
-        def placeholder(*_):
-            """Fail tests more helpfully when execute_module not provided."""
-            raise TypeError(self.__class__.__name__ +
-                            " invoked execute_module without providing the method at initialization.")
-        self.execute_module = execute_module or placeholder
+    def __init__(self, execute_module=None, task_vars=None, tmp=None):
+        self._execute_module = execute_module
+        self.task_vars = task_vars or {}
+        self.tmp = tmp
 
     @abstractproperty
     def name(self):
@@ -42,13 +49,13 @@ class OpenShiftCheck(object):
         """
         return []
 
-    @classmethod
-    def is_active(cls, task_vars):  # pylint: disable=unused-argument
+    @staticmethod
+    def is_active():
         """Returns true if this check applies to the ansible-playbook run."""
         return True
 
     @abstractmethod
-    def run(self, tmp, task_vars):
+    def run(self):
         """Executes a check, normally implemented as a module."""
         return {}
 
@@ -61,6 +68,43 @@ class OpenShiftCheck(object):
             for subclass in subclass.subclasses():
                 yield subclass
 
+    def execute_module(self, module_name=None, module_args=None):
+        """Invoke an Ansible module from a check.
+
+        Invoke stored _execute_module, normally copied from the action
+        plugin, with its params and the task_vars and tmp given at
+        check initialization. No positional parameters beyond these
+        are specified. If it's necessary to specify any of the other
+        parameters to _execute_module then that should just be invoked
+        directly (with awareness of changes in method signature per
+        Ansible version).
+
+        So e.g. check.execute_module("foo", dict(arg1=...))
+        Return: result hash from module execution.
+        """
+        if self._execute_module is None:
+            raise NotImplementedError(
+                self.__class__.__name__ +
+                " invoked execute_module without providing the method at initialization."
+            )
+        return self._execute_module(module_name, module_args, self.tmp, self.task_vars)
+
+    def get_var(self, *keys, **kwargs):
+        """Get deeply nested values from task_vars.
+
+        Ansible task_vars structures are Python dicts, often mapping strings to
+        other dicts. This helper makes it easier to get a nested value, raising
+        OpenShiftCheckException when a key is not found or returning a default value
+        provided as a keyword argument.
+        """
+        try:
+            value = reduce(operator.getitem, keys, self.task_vars)
+        except (KeyError, TypeError):
+            if "default" in kwargs:
+                return kwargs["default"]
+            raise OpenShiftCheckException("'{}' is undefined".format(".".join(map(str, keys))))
+        return value
+
 
 LOADER_EXCLUDES = (
     "__init__.py",
@@ -85,20 +129,3 @@ def load_checks(path=None, subpkg=""):
             modules.append(import_module(__package__ + subpkg + "." + name[:-3]))
 
     return modules
-
-
-def get_var(task_vars, *keys, **kwargs):
-    """Helper function to get deeply nested values from task_vars.
-
-    Ansible task_vars structures are Python dicts, often mapping strings to
-    other dicts. This helper makes it easier to get a nested value, raising
-    OpenShiftCheckException when a key is not found or returning a default value
-    provided as a keyword argument.
-    """
-    try:
-        value = reduce(operator.getitem, keys, task_vars)
-    except (KeyError, TypeError):
-        if "default" in kwargs:
-            return kwargs["default"]
-        raise OpenShiftCheckException("'{}' is undefined".format(".".join(map(str, keys))))
-    return value
diff --git a/roles/openshift_health_checker/openshift_checks/disk_availability.py b/roles/openshift_health_checker/openshift_checks/disk_availability.py
index e93e81efa..283461294 100644
--- a/roles/openshift_health_checker/openshift_checks/disk_availability.py
+++ b/roles/openshift_health_checker/openshift_checks/disk_availability.py
@@ -3,7 +3,7 @@
 import os.path
 import tempfile
 
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 
 
 class DiskAvailability(OpenShiftCheck):
@@ -35,22 +35,21 @@ class DiskAvailability(OpenShiftCheck):
         },
     }
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Skip hosts that do not have recommended disk space requirements."""
-        group_names = get_var(task_vars, "group_names", default=[])
+        group_names = self.get_var("group_names", default=[])
         active_groups = set()
-        for recommendation in cls.recommended_disk_space_bytes.values():
+        for recommendation in self.recommended_disk_space_bytes.values():
             active_groups.update(recommendation.keys())
         has_disk_space_recommendation = bool(active_groups.intersection(group_names))
-        return super(DiskAvailability, cls).is_active(task_vars) and has_disk_space_recommendation
+        return super(DiskAvailability, self).is_active() and has_disk_space_recommendation
 
-    def run(self, tmp, task_vars):
-        group_names = get_var(task_vars, "group_names")
-        ansible_mounts = get_var(task_vars, "ansible_mounts")
+    def run(self):
+        group_names = self.get_var("group_names")
+        ansible_mounts = self.get_var("ansible_mounts")
         ansible_mounts = {mount['mount']: mount for mount in ansible_mounts}
 
-        user_config = get_var(task_vars, "openshift_check_min_host_disk_gb", default={})
+        user_config = self.get_var("openshift_check_min_host_disk_gb", default={})
         try:
             # For backwards-compatibility, if openshift_check_min_host_disk_gb
             # is a number, then it overrides the required config for '/var'.
diff --git a/roles/openshift_health_checker/openshift_checks/docker_image_availability.py b/roles/openshift_health_checker/openshift_checks/docker_image_availability.py
index 0aa11eba1..77180223e 100644
--- a/roles/openshift_health_checker/openshift_checks/docker_image_availability.py
+++ b/roles/openshift_health_checker/openshift_checks/docker_image_availability.py
@@ -1,6 +1,6 @@
 """Check that required Docker images are available."""
 
-from openshift_checks import OpenShiftCheck, get_var
+from openshift_checks import OpenShiftCheck
 from openshift_checks.mixins import DockerHostMixin
 
 
@@ -33,16 +33,15 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
     # to look for images available remotely without waiting to pull them.
     dependencies = ["python-docker-py", "skopeo"]
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Skip hosts with unsupported deployment types."""
-        deployment_type = get_var(task_vars, "openshift_deployment_type")
+        deployment_type = self.get_var("openshift_deployment_type")
         has_valid_deployment_type = deployment_type in DEPLOYMENT_IMAGE_INFO
 
-        return super(DockerImageAvailability, cls).is_active(task_vars) and has_valid_deployment_type
+        return super(DockerImageAvailability, self).is_active() and has_valid_deployment_type
 
-    def run(self, tmp, task_vars):
-        msg, failed, changed = self.ensure_dependencies(task_vars)
+    def run(self):
+        msg, failed, changed = self.ensure_dependencies()
         if failed:
             return {
                 "failed": True,
@@ -50,18 +49,18 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
                 "msg": "Some dependencies are required in order to check Docker image availability.\n" + msg
             }
 
-        required_images = self.required_images(task_vars)
-        missing_images = set(required_images) - set(self.local_images(required_images, task_vars))
+        required_images = self.required_images()
+        missing_images = set(required_images) - set(self.local_images(required_images))
 
         # exit early if all images were found locally
         if not missing_images:
             return {"changed": changed}
 
-        registries = self.known_docker_registries(task_vars)
+        registries = self.known_docker_registries()
         if not registries:
             return {"failed": True, "msg": "Unable to retrieve any docker registries.", "changed": changed}
 
-        available_images = self.available_images(missing_images, registries, task_vars)
+        available_images = self.available_images(missing_images, registries)
         unavailable_images = set(missing_images) - set(available_images)
 
         if unavailable_images:
@@ -76,8 +75,7 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
 
         return {"changed": changed}
 
-    @staticmethod
-    def required_images(task_vars):
+    def required_images(self):
         """
         Determine which images we expect to need for this host.
         Returns: a set of required images like 'openshift/origin:v3.6'
@@ -94,17 +92,17 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
         Registry is not included in constructed images. It may be in oreg_url or etcd image.
         """
         required = set()
-        deployment_type = get_var(task_vars, "openshift_deployment_type")
-        host_groups = get_var(task_vars, "group_names")
+        deployment_type = self.get_var("openshift_deployment_type")
+        host_groups = self.get_var("group_names")
         # containerized etcd may not have openshift_image_tag, see bz 1466622
-        image_tag = get_var(task_vars, "openshift_image_tag", default="latest")
+        image_tag = self.get_var("openshift_image_tag", default="latest")
         image_info = DEPLOYMENT_IMAGE_INFO[deployment_type]
         if not image_info:
             return required
 
         # template for images that run on top of OpenShift
         image_url = "{}/{}-{}:{}".format(image_info["namespace"], image_info["name"], "${component}", "${version}")
-        image_url = get_var(task_vars, "oreg_url", default="") or image_url
+        image_url = self.get_var("oreg_url", default="") or image_url
         if 'nodes' in host_groups:
             for suffix in NODE_IMAGE_SUFFIXES:
                 required.add(image_url.replace("${component}", suffix).replace("${version}", image_tag))
@@ -114,7 +112,7 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
             required.add(image_info["registry_console_image"])
 
         # images for containerized components
-        if get_var(task_vars, "openshift", "common", "is_containerized"):
+        if self.get_var("openshift", "common", "is_containerized"):
             components = set()
             if 'nodes' in host_groups:
                 components.update(["node", "openvswitch"])
@@ -127,28 +125,27 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
 
         return required
 
-    def local_images(self, images, task_vars):
+    def local_images(self, images):
         """Filter a list of images and return those available locally."""
         return [
             image for image in images
-            if self.is_image_local(image, task_vars)
+            if self.is_image_local(image)
         ]
 
-    def is_image_local(self, image, task_vars):
+    def is_image_local(self, image):
         """Check if image is already in local docker index."""
-        result = self.execute_module("docker_image_facts", {"name": image}, task_vars=task_vars)
+        result = self.execute_module("docker_image_facts", {"name": image})
         if result.get("failed", False):
             return False
 
         return bool(result.get("images", []))
 
-    @staticmethod
-    def known_docker_registries(task_vars):
+    def known_docker_registries(self):
         """Build a list of docker registries available according to inventory vars."""
-        docker_facts = get_var(task_vars, "openshift", "docker")
+        docker_facts = self.get_var("openshift", "docker")
         regs = set(docker_facts["additional_registries"])
 
-        deployment_type = get_var(task_vars, "openshift_deployment_type")
+        deployment_type = self.get_var("openshift_deployment_type")
         if deployment_type == "origin":
             regs.update(["docker.io"])
         elif "enterprise" in deployment_type:
@@ -156,14 +153,14 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
 
         return list(regs)
 
-    def available_images(self, images, default_registries, task_vars):
+    def available_images(self, images, default_registries):
         """Search remotely for images. Returns: list of images found."""
         return [
             image for image in images
-            if self.is_available_skopeo_image(image, default_registries, task_vars)
+            if self.is_available_skopeo_image(image, default_registries)
         ]
 
-    def is_available_skopeo_image(self, image, default_registries, task_vars):
+    def is_available_skopeo_image(self, image, default_registries):
         """Use Skopeo to determine if required image exists in known registry(s)."""
         registries = default_registries
 
@@ -174,7 +171,7 @@ class DockerImageAvailability(DockerHostMixin, OpenShiftCheck):
 
         for registry in registries:
             args = {"_raw_params": "skopeo inspect --tls-verify=false docker://{}/{}".format(registry, image)}
-            result = self.execute_module("command", args, task_vars=task_vars)
+            result = self.execute_module("command", args)
             if result.get("rc", 0) == 0 and not result.get("failed"):
                 return True
 
diff --git a/roles/openshift_health_checker/openshift_checks/docker_storage.py b/roles/openshift_health_checker/openshift_checks/docker_storage.py
index d2227d244..dea15a56e 100644
--- a/roles/openshift_health_checker/openshift_checks/docker_storage.py
+++ b/roles/openshift_health_checker/openshift_checks/docker_storage.py
@@ -2,7 +2,7 @@
 import json
 import os.path
 import re
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 from openshift_checks.mixins import DockerHostMixin
 
 
@@ -42,8 +42,8 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
         ),
     ]
 
-    def run(self, tmp, task_vars):
-        msg, failed, changed = self.ensure_dependencies(task_vars)
+    def run(self):
+        msg, failed, changed = self.ensure_dependencies()
         if failed:
             return {
                 "failed": True,
@@ -52,7 +52,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
             }
 
         # attempt to get the docker info hash from the API
-        docker_info = self.execute_module("docker_info", {}, task_vars=task_vars)
+        docker_info = self.execute_module("docker_info", {})
         if docker_info.get("failed"):
             return {"failed": True, "changed": changed,
                     "msg": "Failed to query Docker API. Is docker running on this host?"}
@@ -76,15 +76,15 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
         result = {}
 
         if driver == "devicemapper":
-            result = self.check_devicemapper_support(driver_status, task_vars)
+            result = self.check_devicemapper_support(driver_status)
 
         if driver in ['overlay', 'overlay2']:
-            result = self.check_overlay_support(docker_info, driver_status, task_vars)
+            result = self.check_overlay_support(docker_info, driver_status)
 
         result['changed'] = result.get('changed', False) or changed
         return result
 
-    def check_devicemapper_support(self, driver_status, task_vars):
+    def check_devicemapper_support(self, driver_status):
         """Check if dm storage driver is supported as configured. Return: result dict."""
         if driver_status.get("Data loop file"):
             msg = (
@@ -94,10 +94,10 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
                 "See http://red.ht/2rNperO for further information."
             )
             return {"failed": True, "msg": msg}
-        result = self.check_dm_usage(driver_status, task_vars)
+        result = self.check_dm_usage(driver_status)
         return result
 
-    def check_dm_usage(self, driver_status, task_vars):
+    def check_dm_usage(self, driver_status):
         """Check usage thresholds for Docker dm storage driver. Return: result dict.
         Backing assumptions: We expect devicemapper to be backed by an auto-expanding thin pool
         implemented as an LV in an LVM2 VG. This is how docker-storage-setup currently configures
@@ -109,7 +109,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
         could run out of space first; so we check both.
         """
         vals = dict(
-            vg_free=self.get_vg_free(driver_status.get("Pool Name"), task_vars),
+            vg_free=self.get_vg_free(driver_status.get("Pool Name")),
             data_used=driver_status.get("Data Space Used"),
             data_total=driver_status.get("Data Space Total"),
             metadata_used=driver_status.get("Metadata Space Used"),
@@ -130,7 +130,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
         # determine the threshold percentages which usage should not exceed
         for name, default in [("data", self.max_thinpool_data_usage_percent),
                               ("metadata", self.max_thinpool_meta_usage_percent)]:
-            percent = get_var(task_vars, "max_thinpool_" + name + "_usage_percent", default=default)
+            percent = self.get_var("max_thinpool_" + name + "_usage_percent", default=default)
             try:
                 vals[name + "_threshold"] = float(percent)
             except ValueError:
@@ -157,7 +157,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
         vals["msg"] = "\n".join(messages or ["Thinpool usage is within thresholds."])
         return vals
 
-    def get_vg_free(self, pool, task_vars):
+    def get_vg_free(self, pool):
         """Determine which VG to examine according to the pool name. Return: size vgs reports.
         Pool name is the only indicator currently available from the Docker API driver info.
         We assume a name that looks like "vg--name-docker--pool";
@@ -174,7 +174,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
         vgs_cmd = "/sbin/vgs --noheadings -o vg_free --units g --select vg_name=" + vg_name
         # should return free space like "  12.00g" if the VG exists; empty if it does not
 
-        ret = self.execute_module("command", {"_raw_params": vgs_cmd}, task_vars=task_vars)
+        ret = self.execute_module("command", {"_raw_params": vgs_cmd})
         if ret.get("failed") or ret.get("rc", 0) != 0:
             raise OpenShiftCheckException(
                 "Is LVM installed? Failed to run /sbin/vgs "
@@ -213,7 +213,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
 
         return float(number) * multiplier
 
-    def check_overlay_support(self, docker_info, driver_status, task_vars):
+    def check_overlay_support(self, docker_info, driver_status):
         """Check if overlay storage driver is supported for this host. Return: result dict."""
         # check for xfs as backing store
         backing_fs = driver_status.get("Backing Filesystem", "[NONE]")
@@ -239,13 +239,13 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
             # NOTE: we could check for --selinux-enabled here but docker won't even start with
             # that option until it's supported in the kernel so we don't need to.
 
-        return self.check_overlay_usage(docker_info, task_vars)
+        return self.check_overlay_usage(docker_info)
 
-    def check_overlay_usage(self, docker_info, task_vars):
+    def check_overlay_usage(self, docker_info):
         """Check disk usage on OverlayFS backing store volume. Return: result dict."""
         path = docker_info.get("DockerRootDir", "/var/lib/docker") + "/" + docker_info["Driver"]
 
-        threshold = get_var(task_vars, "max_overlay_usage_percent", default=self.max_overlay_usage_percent)
+        threshold = self.get_var("max_overlay_usage_percent", default=self.max_overlay_usage_percent)
         try:
             threshold = float(threshold)
         except ValueError:
@@ -254,7 +254,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
                 "msg": "Specified 'max_overlay_usage_percent' is not a percentage: {}".format(threshold),
             }
 
-        mount = self.find_ansible_mount(path, get_var(task_vars, "ansible_mounts"))
+        mount = self.find_ansible_mount(path, self.get_var("ansible_mounts"))
         try:
             free_bytes = mount['size_available']
             total_bytes = mount['size_total']
diff --git a/roles/openshift_health_checker/openshift_checks/etcd_imagedata_size.py b/roles/openshift_health_checker/openshift_checks/etcd_imagedata_size.py
index 3097728aa..28c38504d 100644
--- a/roles/openshift_health_checker/openshift_checks/etcd_imagedata_size.py
+++ b/roles/openshift_health_checker/openshift_checks/etcd_imagedata_size.py
@@ -2,7 +2,7 @@
 Ansible module for determining if the size of OpenShift image data exceeds a specified limit in an etcd cluster.
 """
 
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 
 
 class EtcdImageDataSize(OpenShiftCheck):
@@ -11,24 +11,25 @@ class EtcdImageDataSize(OpenShiftCheck):
     name = "etcd_imagedata_size"
     tags = ["etcd"]
 
-    def run(self, tmp, task_vars):
-        etcd_mountpath = self._get_etcd_mountpath(get_var(task_vars, "ansible_mounts"))
+    def run(self):
+        etcd_mountpath = self._get_etcd_mountpath(self.get_var("ansible_mounts"))
         etcd_avail_diskspace = etcd_mountpath["size_available"]
         etcd_total_diskspace = etcd_mountpath["size_total"]
 
-        etcd_imagedata_size_limit = get_var(task_vars,
-                                            "etcd_max_image_data_size_bytes",
-                                            default=int(0.5 * float(etcd_total_diskspace - etcd_avail_diskspace)))
+        etcd_imagedata_size_limit = self.get_var(
+            "etcd_max_image_data_size_bytes",
+            default=int(0.5 * float(etcd_total_diskspace - etcd_avail_diskspace))
+        )
 
-        etcd_is_ssl = get_var(task_vars, "openshift", "master", "etcd_use_ssl", default=False)
-        etcd_port = get_var(task_vars, "openshift", "master", "etcd_port", default=2379)
-        etcd_hosts = get_var(task_vars, "openshift", "master", "etcd_hosts")
+        etcd_is_ssl = self.get_var("openshift", "master", "etcd_use_ssl", default=False)
+        etcd_port = self.get_var("openshift", "master", "etcd_port", default=2379)
+        etcd_hosts = self.get_var("openshift", "master", "etcd_hosts")
 
-        config_base = get_var(task_vars, "openshift", "common", "config_base")
+        config_base = self.get_var("openshift", "common", "config_base")
 
-        cert = task_vars.get("etcd_client_cert", config_base + "/master/master.etcd-client.crt")
-        key = task_vars.get("etcd_client_key", config_base + "/master/master.etcd-client.key")
-        ca_cert = task_vars.get("etcd_client_ca_cert", config_base + "/master/master.etcd-ca.crt")
+        cert = self.get_var("etcd_client_cert", default=config_base + "/master/master.etcd-client.crt")
+        key = self.get_var("etcd_client_key", default=config_base + "/master/master.etcd-client.key")
+        ca_cert = self.get_var("etcd_client_ca_cert", default=config_base + "/master/master.etcd-ca.crt")
 
         for etcd_host in list(etcd_hosts):
             args = {
@@ -46,7 +47,7 @@ class EtcdImageDataSize(OpenShiftCheck):
                 },
             }
 
-            etcdkeysize = self.execute_module("etcdkeysize", args, task_vars)
+            etcdkeysize = self.execute_module("etcdkeysize", args)
 
             if etcdkeysize.get("rc", 0) != 0 or etcdkeysize.get("failed"):
                 msg = 'Failed to retrieve stats for etcd host "{host}": {reason}'
diff --git a/roles/openshift_health_checker/openshift_checks/etcd_traffic.py b/roles/openshift_health_checker/openshift_checks/etcd_traffic.py
index 40c87873d..cc1b14d8a 100644
--- a/roles/openshift_health_checker/openshift_checks/etcd_traffic.py
+++ b/roles/openshift_health_checker/openshift_checks/etcd_traffic.py
@@ -1,6 +1,6 @@
 """Check that scans journalctl for messages caused as a symptom of increased etcd traffic."""
 
-from openshift_checks import OpenShiftCheck, get_var
+from openshift_checks import OpenShiftCheck
 
 
 class EtcdTraffic(OpenShiftCheck):
@@ -9,19 +9,18 @@ class EtcdTraffic(OpenShiftCheck):
     name = "etcd_traffic"
     tags = ["health", "etcd"]
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Skip hosts that do not have etcd in their group names."""
-        group_names = get_var(task_vars, "group_names", default=[])
+        group_names = self.get_var("group_names", default=[])
         valid_group_names = "etcd" in group_names
 
-        version = get_var(task_vars, "openshift", "common", "short_version")
+        version = self.get_var("openshift", "common", "short_version")
         valid_version = version in ("3.4", "3.5", "1.4", "1.5")
 
-        return super(EtcdTraffic, cls).is_active(task_vars) and valid_group_names and valid_version
+        return super(EtcdTraffic, self).is_active() and valid_group_names and valid_version
 
-    def run(self, tmp, task_vars):
-        is_containerized = get_var(task_vars, "openshift", "common", "is_containerized")
+    def run(self):
+        is_containerized = self.get_var("openshift", "common", "is_containerized")
         unit = "etcd_container" if is_containerized else "etcd"
 
         log_matchers = [{
@@ -30,9 +29,7 @@ class EtcdTraffic(OpenShiftCheck):
             "unit": unit
         }]
 
-        match = self.execute_module("search_journalctl", {
-            "log_matchers": log_matchers,
-        }, task_vars)
+        match = self.execute_module("search_journalctl", {"log_matchers": log_matchers})
 
         if match.get("matched"):
             msg = ("Higher than normal etcd traffic detected.\n"
diff --git a/roles/openshift_health_checker/openshift_checks/etcd_volume.py b/roles/openshift_health_checker/openshift_checks/etcd_volume.py
index 7452c9cc1..da7d0364a 100644
--- a/roles/openshift_health_checker/openshift_checks/etcd_volume.py
+++ b/roles/openshift_health_checker/openshift_checks/etcd_volume.py
@@ -1,6 +1,6 @@
 """A health check for OpenShift clusters."""
 
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 
 
 class EtcdVolume(OpenShiftCheck):
@@ -14,21 +14,18 @@ class EtcdVolume(OpenShiftCheck):
     # Where to find ectd data, higher priority first.
     supported_mount_paths = ["/var/lib/etcd", "/var/lib", "/var", "/"]
 
-    @classmethod
-    def is_active(cls, task_vars):
-        etcd_hosts = get_var(task_vars, "groups", "etcd", default=[]) or get_var(task_vars, "groups", "masters",
-                                                                                 default=[]) or []
-        is_etcd_host = get_var(task_vars, "ansible_ssh_host") in etcd_hosts
-        return super(EtcdVolume, cls).is_active(task_vars) and is_etcd_host
+    def is_active(self):
+        etcd_hosts = self.get_var("groups", "etcd", default=[]) or self.get_var("groups", "masters", default=[]) or []
+        is_etcd_host = self.get_var("ansible_ssh_host") in etcd_hosts
+        return super(EtcdVolume, self).is_active() and is_etcd_host
 
-    def run(self, tmp, task_vars):
-        mount_info = self._etcd_mount_info(task_vars)
+    def run(self):
+        mount_info = self._etcd_mount_info()
         available = mount_info["size_available"]
         total = mount_info["size_total"]
         used = total - available
 
-        threshold = get_var(
-            task_vars,
+        threshold = self.get_var(
             "etcd_device_usage_threshold_percent",
             default=self.default_threshold_percent
         )
@@ -45,8 +42,8 @@ class EtcdVolume(OpenShiftCheck):
 
         return {"changed": False}
 
-    def _etcd_mount_info(self, task_vars):
-        ansible_mounts = get_var(task_vars, "ansible_mounts")
+    def _etcd_mount_info(self):
+        ansible_mounts = self.get_var("ansible_mounts")
         mounts = {mnt.get("mount"): mnt for mnt in ansible_mounts}
 
         for path in self.supported_mount_paths:
diff --git a/roles/openshift_health_checker/openshift_checks/logging/curator.py b/roles/openshift_health_checker/openshift_checks/logging/curator.py
index 7e932cf98..f82ae64d7 100644
--- a/roles/openshift_health_checker/openshift_checks/logging/curator.py
+++ b/roles/openshift_health_checker/openshift_checks/logging/curator.py
@@ -1,6 +1,5 @@
 """Check for an aggregated logging Curator deployment"""
 
-from openshift_checks import get_var
 from openshift_checks.logging.logging import LoggingCheck
 
 
@@ -12,13 +11,11 @@ class Curator(LoggingCheck):
 
     logging_namespace = None
 
-    def run(self, tmp, task_vars):
-        self.logging_namespace = get_var(task_vars, "openshift_logging_namespace", default="logging")
+    def run(self):
+        self.logging_namespace = self.get_var("openshift_logging_namespace", default="logging")
         curator_pods, error = super(Curator, self).get_pods_for_component(
-            self.execute_module,
             self.logging_namespace,
             "curator",
-            task_vars
         )
         if error:
             return {"failed": True, "changed": False, "msg": error}
diff --git a/roles/openshift_health_checker/openshift_checks/logging/elasticsearch.py b/roles/openshift_health_checker/openshift_checks/logging/elasticsearch.py
index d57cb9376..1e478c04d 100644
--- a/roles/openshift_health_checker/openshift_checks/logging/elasticsearch.py
+++ b/roles/openshift_health_checker/openshift_checks/logging/elasticsearch.py
@@ -3,7 +3,6 @@
 import json
 import re
 
-from openshift_checks import get_var
 from openshift_checks.logging.logging import LoggingCheck
 
 
@@ -15,19 +14,17 @@ class Elasticsearch(LoggingCheck):
 
     logging_namespace = None
 
-    def run(self, tmp, task_vars):
+    def run(self):
         """Check various things and gather errors. Returns: result as hash"""
 
-        self.logging_namespace = get_var(task_vars, "openshift_logging_namespace", default="logging")
+        self.logging_namespace = self.get_var("openshift_logging_namespace", default="logging")
         es_pods, error = super(Elasticsearch, self).get_pods_for_component(
-            self.execute_module,
             self.logging_namespace,
             "es",
-            task_vars,
         )
         if error:
             return {"failed": True, "changed": False, "msg": error}
-        check_error = self.check_elasticsearch(es_pods, task_vars)
+        check_error = self.check_elasticsearch(es_pods)
 
         if check_error:
             msg = ("The following Elasticsearch deployment issue was found:"
@@ -52,7 +49,7 @@ class Elasticsearch(LoggingCheck):
             ))]
         return not_running, []
 
-    def check_elasticsearch(self, es_pods, task_vars):
+    def check_elasticsearch(self, es_pods):
         """Various checks for elasticsearch. Returns: error string"""
         not_running_pods, error_msgs = self._not_running_elasticsearch_pods(es_pods)
         running_pods = [pod for pod in es_pods if pod not in not_running_pods]
@@ -63,10 +60,10 @@ class Elasticsearch(LoggingCheck):
         }
         if not pods_by_name:
             return 'No logging Elasticsearch pods were found. Is logging deployed?'
-        error_msgs += self._check_elasticsearch_masters(pods_by_name, task_vars)
-        error_msgs += self._check_elasticsearch_node_list(pods_by_name, task_vars)
-        error_msgs += self._check_es_cluster_health(pods_by_name, task_vars)
-        error_msgs += self._check_elasticsearch_diskspace(pods_by_name, task_vars)
+        error_msgs += self._check_elasticsearch_masters(pods_by_name)
+        error_msgs += self._check_elasticsearch_node_list(pods_by_name)
+        error_msgs += self._check_es_cluster_health(pods_by_name)
+        error_msgs += self._check_elasticsearch_diskspace(pods_by_name)
         return '\n'.join(error_msgs)
 
     @staticmethod
@@ -74,14 +71,14 @@ class Elasticsearch(LoggingCheck):
         base = "exec {name} -- curl -s --cert {base}cert --key {base}key --cacert {base}ca -XGET '{url}'"
         return base.format(base="/etc/elasticsearch/secret/admin-", name=pod_name, url=url)
 
-    def _check_elasticsearch_masters(self, pods_by_name, task_vars):
+    def _check_elasticsearch_masters(self, pods_by_name):
         """Check that Elasticsearch masters are sane. Returns: list of error strings"""
         es_master_names = set()
         error_msgs = []
         for pod_name in pods_by_name.keys():
             # Compare what each ES node reports as master and compare for split brain
             get_master_cmd = self._build_es_curl_cmd(pod_name, "https://localhost:9200/_cat/master")
-            master_name_str = self._exec_oc(get_master_cmd, [], task_vars)
+            master_name_str = self._exec_oc(get_master_cmd, [])
             master_names = (master_name_str or '').split(' ')
             if len(master_names) > 1:
                 es_master_names.add(master_names[1])
@@ -106,7 +103,7 @@ class Elasticsearch(LoggingCheck):
 
         return error_msgs
 
-    def _check_elasticsearch_node_list(self, pods_by_name, task_vars):
+    def _check_elasticsearch_node_list(self, pods_by_name):
         """Check that reported ES masters are accounted for by pods. Returns: list of error strings"""
 
         if not pods_by_name:
@@ -114,7 +111,7 @@ class Elasticsearch(LoggingCheck):
 
         # get ES cluster nodes
         node_cmd = self._build_es_curl_cmd(list(pods_by_name.keys())[0], 'https://localhost:9200/_nodes')
-        cluster_node_data = self._exec_oc(node_cmd, [], task_vars)
+        cluster_node_data = self._exec_oc(node_cmd, [])
         try:
             cluster_nodes = json.loads(cluster_node_data)['nodes']
         except (ValueError, KeyError):
@@ -136,12 +133,12 @@ class Elasticsearch(LoggingCheck):
 
         return error_msgs
 
-    def _check_es_cluster_health(self, pods_by_name, task_vars):
+    def _check_es_cluster_health(self, pods_by_name):
         """Exec into the elasticsearch pods and check the cluster health. Returns: list of errors"""
         error_msgs = []
         for pod_name in pods_by_name.keys():
             cluster_health_cmd = self._build_es_curl_cmd(pod_name, 'https://localhost:9200/_cluster/health?pretty=true')
-            cluster_health_data = self._exec_oc(cluster_health_cmd, [], task_vars)
+            cluster_health_data = self._exec_oc(cluster_health_cmd, [])
             try:
                 health_res = json.loads(cluster_health_data)
                 if not health_res or not health_res.get('status'):
@@ -160,7 +157,7 @@ class Elasticsearch(LoggingCheck):
 
         return error_msgs
 
-    def _check_elasticsearch_diskspace(self, pods_by_name, task_vars):
+    def _check_elasticsearch_diskspace(self, pods_by_name):
         """
         Exec into an ES pod and query the diskspace on the persistent volume.
         Returns: list of errors
@@ -168,7 +165,7 @@ class Elasticsearch(LoggingCheck):
         error_msgs = []
         for pod_name in pods_by_name.keys():
             df_cmd = 'exec {} -- df --output=ipcent,pcent /elasticsearch/persistent'.format(pod_name)
-            disk_output = self._exec_oc(df_cmd, [], task_vars)
+            disk_output = self._exec_oc(df_cmd, [])
             lines = disk_output.splitlines()
             # expecting one header looking like 'IUse% Use%' and one body line
             body_re = r'\s*(\d+)%?\s+(\d+)%?\s*$'
@@ -180,7 +177,7 @@ class Elasticsearch(LoggingCheck):
                 continue
             inode_pct, disk_pct = re.match(body_re, lines[1]).groups()
 
-            inode_pct_thresh = get_var(task_vars, 'openshift_check_efk_es_inode_pct', default='90')
+            inode_pct_thresh = self.get_var('openshift_check_efk_es_inode_pct', default='90')
             if int(inode_pct) >= int(inode_pct_thresh):
                 error_msgs.append(
                     'Inode percent usage on the storage volume for logging ES pod "{pod}"\n'
@@ -191,7 +188,7 @@ class Elasticsearch(LoggingCheck):
                         limit=str(inode_pct_thresh),
                         param='openshift_check_efk_es_inode_pct',
                     ))
-            disk_pct_thresh = get_var(task_vars, 'openshift_check_efk_es_storage_pct', default='80')
+            disk_pct_thresh = self.get_var('openshift_check_efk_es_storage_pct', default='80')
             if int(disk_pct) >= int(disk_pct_thresh):
                 error_msgs.append(
                     'Disk percent usage on the storage volume for logging ES pod "{pod}"\n'
@@ -205,11 +202,9 @@ class Elasticsearch(LoggingCheck):
 
         return error_msgs
 
-    def _exec_oc(self, cmd_str, extra_args, task_vars):
+    def _exec_oc(self, cmd_str, extra_args):
         return super(Elasticsearch, self).exec_oc(
-            self.execute_module,
             self.logging_namespace,
             cmd_str,
             extra_args,
-            task_vars,
         )
diff --git a/roles/openshift_health_checker/openshift_checks/logging/fluentd.py b/roles/openshift_health_checker/openshift_checks/logging/fluentd.py
index 1e1e7f2bd..063e707a9 100644
--- a/roles/openshift_health_checker/openshift_checks/logging/fluentd.py
+++ b/roles/openshift_health_checker/openshift_checks/logging/fluentd.py
@@ -2,7 +2,6 @@
 
 import json
 
-from openshift_checks import get_var
 from openshift_checks.logging.logging import LoggingCheck
 
 
@@ -14,19 +13,17 @@ class Fluentd(LoggingCheck):
 
     logging_namespace = None
 
-    def run(self, tmp, task_vars):
+    def run(self):
         """Check various things and gather errors. Returns: result as hash"""
 
-        self.logging_namespace = get_var(task_vars, "openshift_logging_namespace", default="logging")
+        self.logging_namespace = self.get_var("openshift_logging_namespace", default="logging")
         fluentd_pods, error = super(Fluentd, self).get_pods_for_component(
-            self.execute_module,
             self.logging_namespace,
             "fluentd",
-            task_vars,
         )
         if error:
             return {"failed": True, "changed": False, "msg": error}
-        check_error = self.check_fluentd(fluentd_pods, task_vars)
+        check_error = self.check_fluentd(fluentd_pods)
 
         if check_error:
             msg = ("The following Fluentd deployment issue was found:"
@@ -52,10 +49,9 @@ class Fluentd(LoggingCheck):
             ).format(label=node_selector)
         return fluentd_nodes, None
 
-    @staticmethod
-    def _check_node_labeling(nodes_by_name, fluentd_nodes, node_selector, task_vars):
+    def _check_node_labeling(self, nodes_by_name, fluentd_nodes, node_selector):
         """Note if nodes are not labeled as expected. Returns: error string"""
-        intended_nodes = get_var(task_vars, 'openshift_logging_fluentd_hosts', default=['--all'])
+        intended_nodes = self.get_var('openshift_logging_fluentd_hosts', default=['--all'])
         if not intended_nodes or '--all' in intended_nodes:
             intended_nodes = nodes_by_name.keys()
         nodes_missing_labels = set(intended_nodes) - set(fluentd_nodes.keys())
@@ -113,13 +109,15 @@ class Fluentd(LoggingCheck):
             ))
         return None
 
-    def check_fluentd(self, pods, task_vars):
+    def check_fluentd(self, pods):
         """Verify fluentd is running everywhere. Returns: error string"""
 
-        node_selector = get_var(task_vars, 'openshift_logging_fluentd_nodeselector',
-                                default='logging-infra-fluentd=true')
+        node_selector = self.get_var(
+            'openshift_logging_fluentd_nodeselector',
+            default='logging-infra-fluentd=true'
+        )
 
-        nodes_by_name, error = self.get_nodes_by_name(task_vars)
+        nodes_by_name, error = self.get_nodes_by_name()
 
         if error:
             return error
@@ -128,7 +126,7 @@ class Fluentd(LoggingCheck):
             return error
 
         error_msgs = []
-        error = self._check_node_labeling(nodes_by_name, fluentd_nodes, node_selector, task_vars)
+        error = self._check_node_labeling(nodes_by_name, fluentd_nodes, node_selector)
         if error:
             error_msgs.append(error)
         error = self._check_nodes_have_fluentd(pods, fluentd_nodes)
@@ -147,9 +145,9 @@ class Fluentd(LoggingCheck):
 
         return '\n'.join(error_msgs)
 
-    def get_nodes_by_name(self, task_vars):
+    def get_nodes_by_name(self):
         """Retrieve all the node definitions. Returns: dict(name: node), error string"""
-        nodes_json = self._exec_oc("get nodes -o json", [], task_vars)
+        nodes_json = self._exec_oc("get nodes -o json", [])
         try:
             nodes = json.loads(nodes_json)
         except ValueError:  # no valid json - should not happen
@@ -161,9 +159,9 @@ class Fluentd(LoggingCheck):
             for node in nodes['items']
         }, None
 
-    def _exec_oc(self, cmd_str, extra_args, task_vars):
-        return super(Fluentd, self).exec_oc(self.execute_module,
-                                            self.logging_namespace,
-                                            cmd_str,
-                                            extra_args,
-                                            task_vars)
+    def _exec_oc(self, cmd_str, extra_args):
+        return super(Fluentd, self).exec_oc(
+            self.logging_namespace,
+            cmd_str,
+            extra_args,
+        )
diff --git a/roles/openshift_health_checker/openshift_checks/logging/kibana.py b/roles/openshift_health_checker/openshift_checks/logging/kibana.py
index 551e8dfa0..60f94e106 100644
--- a/roles/openshift_health_checker/openshift_checks/logging/kibana.py
+++ b/roles/openshift_health_checker/openshift_checks/logging/kibana.py
@@ -12,7 +12,6 @@ except ImportError:
     from urllib.error import HTTPError, URLError
     import urllib.request as urllib2
 
-from openshift_checks import get_var
 from openshift_checks.logging.logging import LoggingCheck
 
 
@@ -24,22 +23,20 @@ class Kibana(LoggingCheck):
 
     logging_namespace = None
 
-    def run(self, tmp, task_vars):
+    def run(self):
         """Check various things and gather errors. Returns: result as hash"""
 
-        self.logging_namespace = get_var(task_vars, "openshift_logging_namespace", default="logging")
+        self.logging_namespace = self.get_var("openshift_logging_namespace", default="logging")
         kibana_pods, error = super(Kibana, self).get_pods_for_component(
-            self.execute_module,
             self.logging_namespace,
             "kibana",
-            task_vars,
         )
         if error:
             return {"failed": True, "changed": False, "msg": error}
         check_error = self.check_kibana(kibana_pods)
 
         if not check_error:
-            check_error = self._check_kibana_route(task_vars)
+            check_error = self._check_kibana_route()
 
         if check_error:
             msg = ("The following Kibana deployment issue was found:"
@@ -50,7 +47,7 @@ class Kibana(LoggingCheck):
         # TODO(lmeyer): run it all again for the ops cluster
         return {"failed": False, "changed": False, "msg": 'No problems found with Kibana deployment.'}
 
-    def _verify_url_internal(self, url, task_vars):
+    def _verify_url_internal(self, url):
         """
         Try to reach a URL from the host.
         Returns: success (bool), reason (for failure)
@@ -62,7 +59,7 @@ class Kibana(LoggingCheck):
             # TODO(lmeyer): give users option to validate certs
             status_code=302,
         )
-        result = self.execute_module('uri', args, None, task_vars)
+        result = self.execute_module('uri', args)
         if result.get('failed'):
             return result['msg']
         return None
@@ -114,14 +111,14 @@ class Kibana(LoggingCheck):
 
         return None
 
-    def _get_kibana_url(self, task_vars):
+    def _get_kibana_url(self):
         """
         Get kibana route or report error.
         Returns: url (or empty), reason for failure
         """
 
         # Get logging url
-        get_route = self._exec_oc("get route logging-kibana -o json", [], task_vars)
+        get_route = self._exec_oc("get route logging-kibana -o json", [])
         if not get_route:
             return None, 'no_route_exists'
 
@@ -139,7 +136,7 @@ class Kibana(LoggingCheck):
 
         return 'https://{}/'.format(host), None
 
-    def _check_kibana_route(self, task_vars):
+    def _check_kibana_route(self):
         """
         Check to see if kibana route is up and working.
         Returns: error string
@@ -160,12 +157,12 @@ class Kibana(LoggingCheck):
             ),
         )
 
-        kibana_url, error = self._get_kibana_url(task_vars)
+        kibana_url, error = self._get_kibana_url()
         if not kibana_url:
             return known_errors.get(error, error)
 
         # first, check that kibana is reachable from the master.
-        error = self._verify_url_internal(kibana_url, task_vars)
+        error = self._verify_url_internal(kibana_url)
         if error:
             if 'urlopen error [Errno 111] Connection refused' in error:
                 error = (
@@ -190,7 +187,7 @@ class Kibana(LoggingCheck):
 
         # in production we would like the kibana route to work from outside the
         # cluster too; but that may not be the case, so allow disabling just this part.
-        if not get_var(task_vars, "openshift_check_efk_kibana_external", default=True):
+        if not self.get_var("openshift_check_efk_kibana_external", default=True):
             return None
         error = self._verify_url_external(kibana_url)
         if error:
@@ -221,9 +218,9 @@ class Kibana(LoggingCheck):
             return error
         return None
 
-    def _exec_oc(self, cmd_str, extra_args, task_vars):
-        return super(Kibana, self).exec_oc(self.execute_module,
-                                           self.logging_namespace,
-                                           cmd_str,
-                                           extra_args,
-                                           task_vars)
+    def _exec_oc(self, cmd_str, extra_args):
+        return super(Kibana, self).exec_oc(
+            self.logging_namespace,
+            cmd_str,
+            extra_args,
+        )
diff --git a/roles/openshift_health_checker/openshift_checks/logging/logging.py b/roles/openshift_health_checker/openshift_checks/logging/logging.py
index 46fd4f5c7..a48e1c728 100644
--- a/roles/openshift_health_checker/openshift_checks/logging/logging.py
+++ b/roles/openshift_health_checker/openshift_checks/logging/logging.py
@@ -5,7 +5,7 @@ Util functions for performing checks on an Elasticsearch, Fluentd, and Kibana st
 import json
 import os
 
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 
 
 class LoggingCheck(OpenShiftCheck):
@@ -14,31 +14,27 @@ class LoggingCheck(OpenShiftCheck):
     name = "logging"
     logging_namespace = "logging"
 
-    @classmethod
-    def is_active(cls, task_vars):
-        logging_deployed = get_var(task_vars, "openshift_hosted_logging_deploy", default=False)
-        return super(LoggingCheck, cls).is_active(task_vars) and cls.is_first_master(task_vars) and logging_deployed
+    def is_active(self):
+        logging_deployed = self.get_var("openshift_hosted_logging_deploy", default=False)
+        return logging_deployed and super(LoggingCheck, self).is_active() and self.is_first_master()
 
-    @staticmethod
-    def is_first_master(task_vars):
-        """Run only on first master. Returns: bool"""
+    def is_first_master(self):
+        """Determine if running on first master. Returns: bool"""
         # Note: It would be nice to use membership in oo_first_master group, however for now it
         # seems best to avoid requiring that setup and just check this is the first master.
-        hostname = get_var(task_vars, "ansible_ssh_host") or [None]
-        masters = get_var(task_vars, "groups", "masters", default=None) or [None]
-        return masters and masters[0] == hostname
+        hostname = self.get_var("ansible_ssh_host") or [None]
+        masters = self.get_var("groups", "masters", default=None) or [None]
+        return masters[0] == hostname
 
-    def run(self, tmp, task_vars):
+    def run(self):
         pass
 
-    def get_pods_for_component(self, execute_module, namespace, logging_component, task_vars):
+    def get_pods_for_component(self, namespace, logging_component):
         """Get all pods for a given component. Returns: list of pods for component, error string"""
         pod_output = self.exec_oc(
-            execute_module,
             namespace,
             "get pods -l component={} -o json".format(logging_component),
             [],
-            task_vars
         )
         try:
             pods = json.loads(pod_output)
@@ -64,14 +60,13 @@ class LoggingCheck(OpenShiftCheck):
             )
         ]
 
-    @staticmethod
-    def exec_oc(execute_module=None, namespace="logging", cmd_str="", extra_args=None, task_vars=None):
+    def exec_oc(self, namespace="logging", cmd_str="", extra_args=None):
         """
         Execute an 'oc' command in the remote host.
         Returns: output of command and namespace,
         or raises OpenShiftCheckException on error
         """
-        config_base = get_var(task_vars, "openshift", "common", "config_base")
+        config_base = self.get_var("openshift", "common", "config_base")
         args = {
             "namespace": namespace,
             "config_file": os.path.join(config_base, "master", "admin.kubeconfig"),
@@ -79,7 +74,7 @@ class LoggingCheck(OpenShiftCheck):
             "extra_args": list(extra_args) if extra_args else [],
         }
 
-        result = execute_module("ocutil", args, None, task_vars)
+        result = self.execute_module("ocutil", args)
         if result.get("failed"):
             msg = (
                 'Unexpected error using `oc` to validate the logging stack components.\n'
diff --git a/roles/openshift_health_checker/openshift_checks/logging/logging_index_time.py b/roles/openshift_health_checker/openshift_checks/logging/logging_index_time.py
index 2ddd7549d..b24e88e05 100644
--- a/roles/openshift_health_checker/openshift_checks/logging/logging_index_time.py
+++ b/roles/openshift_health_checker/openshift_checks/logging/logging_index_time.py
@@ -7,7 +7,7 @@ import time
 
 from uuid import uuid4
 
-from openshift_checks import get_var, OpenShiftCheckException
+from openshift_checks import OpenShiftCheckException
 from openshift_checks.logging.logging import LoggingCheck
 
 
@@ -21,11 +21,11 @@ class LoggingIndexTime(LoggingCheck):
 
     logging_namespace = "logging"
 
-    def run(self, tmp, task_vars):
+    def run(self):
         """Add log entry by making unique request to Kibana. Check for unique entry in the ElasticSearch pod logs."""
         try:
             log_index_timeout = int(
-                get_var(task_vars, "openshift_check_logging_index_timeout_seconds", default=ES_CMD_TIMEOUT_SECONDS)
+                self.get_var("openshift_check_logging_index_timeout_seconds", default=ES_CMD_TIMEOUT_SECONDS)
             )
         except ValueError:
             return {
@@ -37,11 +37,9 @@ class LoggingIndexTime(LoggingCheck):
         running_component_pods = dict()
 
         # get all component pods
-        self.logging_namespace = get_var(task_vars, "openshift_logging_namespace", default=self.logging_namespace)
+        self.logging_namespace = self.get_var("openshift_logging_namespace", default=self.logging_namespace)
         for component, name in (['kibana', 'Kibana'], ['es', 'Elasticsearch']):
-            pods, error = self.get_pods_for_component(
-                self.execute_module, self.logging_namespace, component, task_vars,
-            )
+            pods, error = self.get_pods_for_component(self.logging_namespace, component)
 
             if error:
                 msg = 'Unable to retrieve pods for the {} logging component: {}'
@@ -56,29 +54,29 @@ class LoggingIndexTime(LoggingCheck):
 
             running_component_pods[component] = running_pods
 
-        uuid = self.curl_kibana_with_uuid(running_component_pods["kibana"][0], task_vars)
-        self.wait_until_cmd_or_err(running_component_pods["es"][0], uuid, log_index_timeout, task_vars)
+        uuid = self.curl_kibana_with_uuid(running_component_pods["kibana"][0])
+        self.wait_until_cmd_or_err(running_component_pods["es"][0], uuid, log_index_timeout)
         return {}
 
-    def wait_until_cmd_or_err(self, es_pod, uuid, timeout_secs, task_vars):
+    def wait_until_cmd_or_err(self, es_pod, uuid, timeout_secs):
         """Retry an Elasticsearch query every second until query success, or a defined
         length of time has passed."""
         deadline = time.time() + timeout_secs
         interval = 1
-        while not self.query_es_from_es(es_pod, uuid, task_vars):
+        while not self.query_es_from_es(es_pod, uuid):
             if time.time() + interval > deadline:
                 msg = "expecting match in Elasticsearch for message with uuid {}, but no matches were found after {}s."
                 raise OpenShiftCheckException(msg.format(uuid, timeout_secs))
             time.sleep(interval)
 
-    def curl_kibana_with_uuid(self, kibana_pod, task_vars):
+    def curl_kibana_with_uuid(self, kibana_pod):
         """curl Kibana with a unique uuid."""
         uuid = self.generate_uuid()
         pod_name = kibana_pod["metadata"]["name"]
         exec_cmd = "exec {pod_name} -c kibana -- curl --max-time 30 -s http://localhost:5601/{uuid}"
         exec_cmd = exec_cmd.format(pod_name=pod_name, uuid=uuid)
 
-        error_str = self.exec_oc(self.execute_module, self.logging_namespace, exec_cmd, [], task_vars)
+        error_str = self.exec_oc(self.logging_namespace, exec_cmd, [])
 
         try:
             error_code = json.loads(error_str)["statusCode"]
@@ -97,7 +95,7 @@ class LoggingIndexTime(LoggingCheck):
 
         return uuid
 
-    def query_es_from_es(self, es_pod, uuid, task_vars):
+    def query_es_from_es(self, es_pod, uuid):
         """curl the Elasticsearch pod and look for a unique uuid in its logs."""
         pod_name = es_pod["metadata"]["name"]
         exec_cmd = (
@@ -108,7 +106,7 @@ class LoggingIndexTime(LoggingCheck):
             "https://logging-es:9200/project.{namespace}*/_count?q=message:{uuid}"
         )
         exec_cmd = exec_cmd.format(pod_name=pod_name, namespace=self.logging_namespace, uuid=uuid)
-        result = self.exec_oc(self.execute_module, self.logging_namespace, exec_cmd, [], task_vars)
+        result = self.exec_oc(self.logging_namespace, exec_cmd, [])
 
         try:
             count = json.loads(result)["count"]
diff --git a/roles/openshift_health_checker/openshift_checks/memory_availability.py b/roles/openshift_health_checker/openshift_checks/memory_availability.py
index 2b42c72a9..765ba072d 100644
--- a/roles/openshift_health_checker/openshift_checks/memory_availability.py
+++ b/roles/openshift_health_checker/openshift_checks/memory_availability.py
@@ -1,5 +1,5 @@
 """Check that recommended memory is available."""
-from openshift_checks import OpenShiftCheck, get_var
+from openshift_checks import OpenShiftCheck
 
 MIB = 2**20
 GIB = 2**30
@@ -21,19 +21,18 @@ class MemoryAvailability(OpenShiftCheck):
     # https://access.redhat.com/solutions/3006511 physical RAM is partly reserved from memtotal
     memtotal_adjustment = 1 * GIB
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Skip hosts that do not have recommended memory requirements."""
-        group_names = get_var(task_vars, "group_names", default=[])
-        has_memory_recommendation = bool(set(group_names).intersection(cls.recommended_memory_bytes))
-        return super(MemoryAvailability, cls).is_active(task_vars) and has_memory_recommendation
+        group_names = self.get_var("group_names", default=[])
+        has_memory_recommendation = bool(set(group_names).intersection(self.recommended_memory_bytes))
+        return super(MemoryAvailability, self).is_active() and has_memory_recommendation
 
-    def run(self, tmp, task_vars):
-        group_names = get_var(task_vars, "group_names")
-        total_memory_bytes = get_var(task_vars, "ansible_memtotal_mb") * MIB
+    def run(self):
+        group_names = self.get_var("group_names")
+        total_memory_bytes = self.get_var("ansible_memtotal_mb") * MIB
 
         recommended_min = max(self.recommended_memory_bytes.get(name, 0) for name in group_names)
-        configured_min = float(get_var(task_vars, "openshift_check_min_host_memory_gb", default=0)) * GIB
+        configured_min = float(self.get_var("openshift_check_min_host_memory_gb", default=0)) * GIB
         min_memory_bytes = configured_min or recommended_min
 
         if total_memory_bytes + self.memtotal_adjustment < min_memory_bytes:
diff --git a/roles/openshift_health_checker/openshift_checks/mixins.py b/roles/openshift_health_checker/openshift_checks/mixins.py
index 2cb2e21aa..3b2c64e6a 100644
--- a/roles/openshift_health_checker/openshift_checks/mixins.py
+++ b/roles/openshift_health_checker/openshift_checks/mixins.py
@@ -2,19 +2,16 @@
 Mixin classes meant to be used with subclasses of OpenShiftCheck.
 """
 
-from openshift_checks import get_var
-
 
 class NotContainerizedMixin(object):
     """Mixin for checks that are only active when not in containerized mode."""
     # permanent # pylint: disable=too-few-public-methods
     # Reason: The mixin is not intended to stand on its own as a class.
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Only run on non-containerized hosts."""
-        is_containerized = get_var(task_vars, "openshift", "common", "is_containerized")
-        return super(NotContainerizedMixin, cls).is_active(task_vars) and not is_containerized
+        is_containerized = self.get_var("openshift", "common", "is_containerized")
+        return super(NotContainerizedMixin, self).is_active() and not is_containerized
 
 
 class DockerHostMixin(object):
@@ -22,28 +19,26 @@ class DockerHostMixin(object):
 
     dependencies = []
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Only run on hosts that depend on Docker."""
-        is_containerized = get_var(task_vars, "openshift", "common", "is_containerized")
-        is_node = "nodes" in get_var(task_vars, "group_names", default=[])
-        return super(DockerHostMixin, cls).is_active(task_vars) and (is_containerized or is_node)
+        is_containerized = self.get_var("openshift", "common", "is_containerized")
+        is_node = "nodes" in self.get_var("group_names", default=[])
+        return super(DockerHostMixin, self).is_active() and (is_containerized or is_node)
 
-    def ensure_dependencies(self, task_vars):
+    def ensure_dependencies(self):
         """
         Ensure that docker-related packages exist, but not on atomic hosts
         (which would not be able to install but should already have them).
         Returns: msg, failed, changed
         """
-        if get_var(task_vars, "openshift", "common", "is_atomic"):
+        if self.get_var("openshift", "common", "is_atomic"):
             return "", False, False
 
         # NOTE: we would use the "package" module but it's actually an action plugin
         # and it's not clear how to invoke one of those. This is about the same anyway:
         result = self.execute_module(
-            get_var(task_vars, "ansible_pkg_mgr", default="yum"),
+            self.get_var("ansible_pkg_mgr", default="yum"),
             {"name": self.dependencies, "state": "present"},
-            task_vars=task_vars,
         )
         msg = result.get("msg", "")
         if result.get("failed"):
diff --git a/roles/openshift_health_checker/openshift_checks/ovs_version.py b/roles/openshift_health_checker/openshift_checks/ovs_version.py
index 2dd045f1f..cd6ebd493 100644
--- a/roles/openshift_health_checker/openshift_checks/ovs_version.py
+++ b/roles/openshift_health_checker/openshift_checks/ovs_version.py
@@ -3,7 +3,7 @@ Ansible module for determining if an installed version of Open vSwitch is incomp
 currently installed version of OpenShift.
 """
 
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 from openshift_checks.mixins import NotContainerizedMixin
 
 
@@ -27,27 +27,26 @@ class OvsVersion(NotContainerizedMixin, OpenShiftCheck):
         "1": "3",
     }
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Skip hosts that do not have package requirements."""
-        group_names = get_var(task_vars, "group_names", default=[])
+        group_names = self.get_var("group_names", default=[])
         master_or_node = 'masters' in group_names or 'nodes' in group_names
-        return super(OvsVersion, cls).is_active(task_vars) and master_or_node
+        return super(OvsVersion, self).is_active() and master_or_node
 
-    def run(self, tmp, task_vars):
+    def run(self):
         args = {
             "package_list": [
                 {
                     "name": "openvswitch",
-                    "version": self.get_required_ovs_version(task_vars),
+                    "version": self.get_required_ovs_version(),
                 },
             ],
         }
-        return self.execute_module("rpm_version", args, task_vars=task_vars)
+        return self.execute_module("rpm_version", args)
 
-    def get_required_ovs_version(self, task_vars):
+    def get_required_ovs_version(self):
         """Return the correct Open vSwitch version for the current OpenShift version"""
-        openshift_version = self._get_openshift_version(task_vars)
+        openshift_version = self._get_openshift_version()
 
         if float(openshift_version) < 3.5:
             return self.openshift_to_ovs_version["3.4"]
@@ -59,8 +58,8 @@ class OvsVersion(NotContainerizedMixin, OpenShiftCheck):
         msg = "There is no recommended version of Open vSwitch for the current version of OpenShift: {}"
         raise OpenShiftCheckException(msg.format(openshift_version))
 
-    def _get_openshift_version(self, task_vars):
-        openshift_version = get_var(task_vars, "openshift_image_tag")
+    def _get_openshift_version(self):
+        openshift_version = self.get_var("openshift_image_tag")
         if openshift_version and openshift_version[0] == 'v':
             openshift_version = openshift_version[1:]
 
diff --git a/roles/openshift_health_checker/openshift_checks/package_availability.py b/roles/openshift_health_checker/openshift_checks/package_availability.py
index a67b68d1b..a86180b00 100644
--- a/roles/openshift_health_checker/openshift_checks/package_availability.py
+++ b/roles/openshift_health_checker/openshift_checks/package_availability.py
@@ -1,6 +1,6 @@
 """Check that required RPM packages are available."""
 
-from openshift_checks import OpenShiftCheck, get_var
+from openshift_checks import OpenShiftCheck
 from openshift_checks.mixins import NotContainerizedMixin
 
 
@@ -10,14 +10,13 @@ class PackageAvailability(NotContainerizedMixin, OpenShiftCheck):
     name = "package_availability"
     tags = ["preflight"]
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Run only when yum is the package manager as the code is specific to it."""
-        return super(PackageAvailability, cls).is_active(task_vars) and task_vars["ansible_pkg_mgr"] == "yum"
+        return super(PackageAvailability, self).is_active() and self.get_var("ansible_pkg_mgr") == "yum"
 
-    def run(self, tmp, task_vars):
-        rpm_prefix = get_var(task_vars, "openshift", "common", "service_type")
-        group_names = get_var(task_vars, "group_names", default=[])
+    def run(self):
+        rpm_prefix = self.get_var("openshift", "common", "service_type")
+        group_names = self.get_var("group_names", default=[])
 
         packages = set()
 
@@ -27,7 +26,7 @@ class PackageAvailability(NotContainerizedMixin, OpenShiftCheck):
             packages.update(self.node_packages(rpm_prefix))
 
         args = {"packages": sorted(set(packages))}
-        return self.execute_module("check_yum_update", args, tmp=tmp, task_vars=task_vars)
+        return self.execute_module("check_yum_update", args)
 
     @staticmethod
     def master_packages(rpm_prefix):
diff --git a/roles/openshift_health_checker/openshift_checks/package_update.py b/roles/openshift_health_checker/openshift_checks/package_update.py
index db063158c..1e9aecbe0 100644
--- a/roles/openshift_health_checker/openshift_checks/package_update.py
+++ b/roles/openshift_health_checker/openshift_checks/package_update.py
@@ -9,6 +9,6 @@ class PackageUpdate(NotContainerizedMixin, OpenShiftCheck):
     name = "package_update"
     tags = ["preflight"]
 
-    def run(self, tmp, task_vars):
+    def run(self):
         args = {"packages": []}
-        return self.execute_module("check_yum_update", args, tmp=tmp, task_vars=task_vars)
+        return self.execute_module("check_yum_update", args)
diff --git a/roles/openshift_health_checker/openshift_checks/package_version.py b/roles/openshift_health_checker/openshift_checks/package_version.py
index ab4295770..020786804 100644
--- a/roles/openshift_health_checker/openshift_checks/package_version.py
+++ b/roles/openshift_health_checker/openshift_checks/package_version.py
@@ -1,5 +1,5 @@
 """Check that available RPM packages match the required versions."""
-from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
+from openshift_checks import OpenShiftCheck, OpenShiftCheckException
 from openshift_checks.mixins import NotContainerizedMixin
 
 
@@ -28,29 +28,28 @@ class PackageVersion(NotContainerizedMixin, OpenShiftCheck):
         "1": "3",
     }
 
-    @classmethod
-    def is_active(cls, task_vars):
+    def is_active(self):
         """Skip hosts that do not have package requirements."""
-        group_names = get_var(task_vars, "group_names", default=[])
+        group_names = self.get_var("group_names", default=[])
         master_or_node = 'masters' in group_names or 'nodes' in group_names
-        return super(PackageVersion, cls).is_active(task_vars) and master_or_node
+        return super(PackageVersion, self).is_active() and master_or_node
 
-    def run(self, tmp, task_vars):
-        rpm_prefix = get_var(task_vars, "openshift", "common", "service_type")
-        openshift_release = get_var(task_vars, "openshift_release", default='')
-        deployment_type = get_var(task_vars, "openshift_deployment_type")
+    def run(self):
+        rpm_prefix = self.get_var("openshift", "common", "service_type")
+        openshift_release = self.get_var("openshift_release", default='')
+        deployment_type = self.get_var("openshift_deployment_type")
         check_multi_minor_release = deployment_type in ['openshift-enterprise']
 
         args = {
             "package_list": [
                 {
                     "name": "openvswitch",
-                    "version": self.get_required_ovs_version(task_vars),
+                    "version": self.get_required_ovs_version(),
                     "check_multi": False,
                 },
                 {
                     "name": "docker",
-                    "version": self.get_required_docker_version(task_vars),
+                    "version": self.get_required_docker_version(),
                     "check_multi": False,
                 },
                 {
@@ -71,13 +70,13 @@ class PackageVersion(NotContainerizedMixin, OpenShiftCheck):
             ],
         }
 
-        return self.execute_module("aos_version", args, tmp=tmp, task_vars=task_vars)
+        return self.execute_module("aos_version", args)
 
-    def get_required_ovs_version(self, task_vars):
+    def get_required_ovs_version(self):
         """Return the correct Open vSwitch version for the current OpenShift version.
         If the current OpenShift version is >= 3.5, ensure Open vSwitch version 2.6,
         Else ensure Open vSwitch version 2.4"""
-        openshift_version = self.get_openshift_version(task_vars)
+        openshift_version = self.get_openshift_version()
 
         if float(openshift_version) < 3.5:
             return self.openshift_to_ovs_version["3.4"]
@@ -89,12 +88,12 @@ class PackageVersion(NotContainerizedMixin, OpenShiftCheck):
         msg = "There is no recommended version of Open vSwitch for the current version of OpenShift: {}"
         raise OpenShiftCheckException(msg.format(openshift_version))
 
-    def get_required_docker_version(self, task_vars):
+    def get_required_docker_version(self):
         """Return the correct Docker version for the current OpenShift version.
         If the OpenShift version is 3.1, ensure Docker version 1.8.
         If the OpenShift version is 3.2 or 3.3, ensure Docker version 1.10.
         If the current OpenShift version is >= 3.4, ensure Docker version 1.12."""
-        openshift_version = self.get_openshift_version(task_vars)
+        openshift_version = self.get_openshift_version()
 
         if float(openshift_version) >= 3.4:
             return self.openshift_to_docker_version["3.4"]
@@ -106,9 +105,9 @@ class PackageVersion(NotContainerizedMixin, OpenShiftCheck):
         msg = "There is no recommended version of Docker for the current version of OpenShift: {}"
         raise OpenShiftCheckException(msg.format(openshift_version))
 
-    def get_openshift_version(self, task_vars):
+    def get_openshift_version(self):
         """Return received image tag as a normalized X.Y minor version string."""
-        openshift_version = get_var(task_vars, "openshift_image_tag")
+        openshift_version = self.get_var("openshift_image_tag")
         if openshift_version and openshift_version[0] == 'v':
             openshift_version = openshift_version[1:]