check layout.yaml schema before k8s installation (#5184)

contrib/kubespray/quick-start-kubespray.sh

@@ -7,6 +7,9 @@ CLUSTER_CONFIG="$PWD/config/config.yaml"
 echo "layout config file path: ${LAYOUT}"
 echo "cluster config file path: ${CLUSTER_CONFIG}"
 
+echo "Checking layout.yaml schema..."
+python3 script/validate_layout_schema.py -l ${LAYOUT}
+
 echo "Setting up environment..."
 /bin/bash script/environment.sh -c ${CLUSTER_CONFIG} || exit $?
 

contrib/kubespray/script/environment.sh

@@ -33,11 +33,8 @@ sudo apt-get -y install software-properties-common python3 python3-dev
 curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
 sudo python3 get-pip.py
 
-echo "Install paramiko"
-sudo pip3 install paramiko
-
-echo "Install kubernetes package"
-sudo pip3 install kubernetes
+echo "Install python packages"
+sudo pip3 install paramiko kubernetes schema
 
 echo "Install sshpass"
 sudo apt-get -y install sshpass

contrib/kubespray/script/pre-check-generator.py

@@ -76,7 +76,6 @@ def main():
 
     # fill in cpu, memory, computing_device information in both masters and workers
     # we assume the layout file the user gives is correct
-    # TO DO: check layout file before this step
     all_machines = masters + workers
     for machine in all_machines:
         sku_info = layout['machine-sku'][machine['machine-type']]

contrib/kubespray/script/validate_layout_schema.py

@@ -0,0 +1,123 @@
+import argparse
+import logging
+import logging.config
+import sys
+from schema import Schema, Or, Optional, Regex
+import yaml
+
+
+def setup_logger_config(logger):
+    """
+    Setup logging configuration.
+    """
+    if len(logger.handlers) == 0:
+        logger.propagate = False
+        logger.setLevel(logging.DEBUG)
+        consoleHandler = logging.StreamHandler()
+        consoleHandler.setLevel(logging.DEBUG)
+        formatter = logging.Formatter('%(asctime)s [%(levelname)s] - %(filename)s:%(lineno)s : %(message)s')
+        consoleHandler.setFormatter(formatter)
+        logger.addHandler(consoleHandler)
+
+logger = logging.getLogger(__name__)
+setup_logger_config(logger)
+
+
+def load_yaml_config(config_path):
+    with open(config_path, "r") as f:
+        config_data = yaml.load(f, yaml.SafeLoader)
+    return config_data
+
+
+def validate_layout_schema(layout):
+    schema = Schema(
+        {
+            'machine-sku': {
+                str: {
+                    'mem': str,
+                    'cpu': {
+                        'vcore': int,
+                    },
+                    Optional('computing-device'): {
+                        'type': str,
+                        'model': str,
+                        'count': int,
+                    }
+                }
+            },
+            'machine-list': [
+                {
+                    # https://github.com/kubernetes-sigs/kubespray/blob/release-2.11/roles/kubernetes/preinstall/tasks/0020-verify-settings.yml#L124
+                    'hostname': Regex(r"^[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*$"),
+                    'hostip': Regex(r"^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$"),
+                    'machine-type': str,
+                    Optional('pai-master'): Or('true', 'false'),
+                    Optional('pai-worker'): Or('true', 'false'),
+                    Optional('pai-storage'): Or('true', 'false'),
+                }
+            ]
+        }
+    )
+    return schema.validate(layout)
+
+
+def check_layout(layout):
+    # hostname / hostip should be unique
+    hostnames = [elem['hostname'] for elem in layout['machine-list']]
+    if len(hostnames) != len(set(hostnames)):
+        logger.error("hostname should be unique")
+        return False
+    hostips = [elem['hostip'] for elem in layout['machine-list']]
+    if len(hostips) != len(set(hostips)):
+        logger.error("hostip should be unique")
+        return False
+
+    # machine-type should be defined in machine-sku
+    for machine in layout['machine-list']:
+        if machine['machine-type'] not in layout['machine-sku']:
+            logger.error("machine-type {} is not defined".format(machine['machine-type']))
+            return False
+
+    masters = list(filter(lambda elem: 'pai-master' in elem and elem["pai-master"] == 'true', layout['machine-list']))
+    workers = list(filter(lambda elem: 'pai-worker' in elem and elem["pai-worker"] == 'true', layout['machine-list']))
+    # only one pai-master
+    if len(masters) == 0:
+        logger.error('No master machine specified.')
+        return False
+    if len(masters) > 1:
+        logger.error('More than one master machine specified.')
+        return False
+    # at least one pai-worker
+    if len(workers) == 0:
+        logger.error('No worker machine specified.')
+        return False
+    # pai-master / pai-worker cannot be true at the same time
+    if 'pai-worker' in masters[0] and masters[0]['pai-worker'] == 'true':
+        logger.error("One machine can not be pai-master and pai-worker at the same time.")
+        return False
+
+    return True
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-l', '--layout', dest="layout", required=True,
+                        help="layout.yaml")
+    args = parser.parse_args()
+
+    layout = load_yaml_config(args.layout)
+    try:
+        validate_layout_schema(layout)
+    except Exception as exp:
+        logger.error("layout.yaml schema validation failed: \n %s", exp)
+        sys.exit(1)
+
+    if not check_layout(layout):
+        logger.error("layout.yaml schema validation failed")
+        sys.exit(1)
+
+    logger.info("layout.yaml schema validation succeeded.")
+
+
+if __name__ == "__main__":
+    main()