Add support to ignore checking pods in ns

This commit adds a way for user to specify a regex to ignore checking
the pods in the namespaces.

Fixes https://github.com/redhat-chaos/cerberus/issues/177.

cerberus/kubernetes/client.py

@@ -261,14 +261,21 @@ def namespace_sleep_tracker(namespace, pods_tracker):
 
 # Monitor the status of the pods in the specified namespace
 # and set the status to true or false
-def monitor_namespace(namespace):
+def monitor_namespace(namespace, ignore_pattern=None):
     notready_pods = set()
+    match = False
     notready_containers = defaultdict(list)
     all_pod_info_list = get_all_pod_info(namespace)
     if all_pod_info_list is not None and len(all_pod_info_list) > 0:
         for all_pod_info in all_pod_info_list:
             for pod_info in all_pod_info.items:
                 pod = pod_info.metadata.name
+                if ignore_pattern:
+                    for pattern in ignore_pattern:
+                        if re.match(pattern, pod):
+                            match = True
+                if match:
+                    continue
                 pod_status = pod_info.status
                 pod_status_phase = pod_status.phase
                 if pod_status_phase != "Running" and pod_status_phase != "Succeeded":
@@ -295,8 +302,8 @@ def monitor_namespace(namespace):
     return status, notready_pods, notready_containers
 
 
-def process_namespace(iteration, namespace, failed_pods_components, failed_pod_containers):
-    watch_component_status, failed_component_pods, failed_containers = monitor_namespace(namespace)
+def process_namespace(iteration, namespace, failed_pods_components, failed_pod_containers, ignore_pattern):
+    watch_component_status, failed_component_pods, failed_containers = monitor_namespace(namespace, ignore_pattern)
     logging.info("Iteration %s: %s: %s" % (iteration, namespace, watch_component_status))
     if not watch_component_status:
         failed_pods_components[namespace] = failed_component_pods

config/config.yaml

@@ -19,6 +19,7 @@ cerberus:
         -    openshift-kube-scheduler
         -    openshift-ingress
         -    openshift-sdn                               # When enabled, it will check for the cluster sdn and monitor that namespace
+    watch_namespaces_ignore_pattern: [^installer*]       # Ignores pods matching the regex pattern in the namespaces specified under watch_namespaces
     cerberus_publish_status: True                        # When enabled, cerberus starts a light weight http server and publishes the status
     inspect_components: False                            # Enable it only when OpenShift client is supported to run
                                                          # When enabled, cerberus collects logs, events and metrics of failed components

config/kubernetes_config.yaml

@@ -9,6 +9,7 @@ cerberus:
         label: node-role.kubernetes.io/master
     watch_namespaces:                                    # List of namespaces to be monitored
         -    kube-system
+    watch_namespaces_ignore_pattern: []                  # Ignores pods matching the regex pattern in the namespaces specified under watch_namespaces
     cerberus_publish_status: True                        # When enabled, cerberus starts a light weight http server and publishes the status
     inspect_components: False                            # Enable it only when OpenShift client is supported to run
                                                          # When enabled, cerberus collects logs, events and metrics of failed components

docs/config.md

@@ -37,6 +37,7 @@ cerberus:
         -    openshift-kube-scheduler
         -    openshift-ingress
         -    openshift-sdn                                   # When enabled, it will check for the cluster sdn and monitor that namespace
+    watch_namespaces_ignore_pattern: []                  # Ignores pods matching the regex pattern in the namespaces specified under watch_namespaces
     cerberus_publish_status: True                        # When enabled, cerberus starts a light weight http server and publishes the status
     inspect_components: False                            # Enable it only when OpenShift client is supported to run
                                                          # When enabled, cerberus collects logs, events and metrics of failed components

start_cerberus.py

@@ -79,6 +79,7 @@ def main(cfg):
         watch_nodes = config["cerberus"].get("watch_nodes", False)
         watch_cluster_operators = config["cerberus"].get("watch_cluster_operators", False)
         watch_namespaces = config["cerberus"].get("watch_namespaces", [])
+        watch_namespaces_ignore_pattern = config["cerberus"].get("watch_namespaces_ignore_pattern", [])
         watch_terminating_namespaces = config["cerberus"].get("watch_terminating_namespaces", True)
         watch_url_routes = config["cerberus"].get("watch_url_routes", [])
         watch_master_schedulable = config["cerberus"].get("watch_master_schedulable", {})
@@ -288,6 +289,7 @@ def main(cfg):
                         watch_namespaces,
                         repeat(failed_pods_components),
                         repeat(failed_pod_containers),
+                        repeat(watch_namespaces_ignore_pattern),
                     ),
                 )