Switch to 'logging' module instead of 'syslog'

src/vmpoller-client

@@ -30,16 +30,18 @@ vmpoller-client is an application used for polling objects' information from a V
 It is intended to be integrated into a Zabbix template for polling of ESX hosts and Datastores properties.
 """
 
+import logging
+
 from vmpoller.core import VMPollerClient
 from docopt import docopt
 
 def main():
 
     usage="""
 Usage:
-  vmpoller-client [-r <retries>] [-t <timeout>] (-D|-H) -c discover -V <vcenter> -e <endpoint>
-  vmpoller-client [-r <retries>] [-t <timeout>] -H -n <name> -p <property> -c poll -V <vcenter> -e <endpoint>
-  vmpoller-client [-r <retries>] [-t <timeout>] -D -n <name> -p <property> -u <datastore-url> -c poll -V <vcenter> -e <endpoint>
+  vmpoller-client [-r <retries>] [-t <timeout>] [-l <logfile>] (-D|-H) -c discover -V <vcenter> -e <endpoint>
+  vmpoller-client [-r <retries>] [-t <timeout>] [-l <logfile>] -H -n <name> -p <property> -c poll -V <vcenter> -e <endpoint>
+  vmpoller-client [-r <retries>] [-t <timeout>] [-l <logfile>] -D -n <name> -p <property> -u <datastore-url> -c poll -V <vcenter> -e <endpoint>
   vmpoller-client --help
   vmpoller-client --version
 
@@ -56,11 +58,17 @@ Options:
   -r <retries>, --retries <retries>         Number of time to retry if a request times out [default: 3]
   -t <timeout>, --timeout <timeout>         Timeout after that period of milliseconds [default: 3000]
   -e <endpoint>, --endpoint <endpoint>      Endpoint of ZeroMQ Proxy/Broker the client connects to
+  -l <logfile>, --logfile <logfile>         Specify the log file to use when performing the poll
+                                            [default: /var/log/vmpoller/vmpoller-client.log]
 
 """
 
     args = docopt(usage, version="1.0.0")
 
+    logging.basicConfig(filename=args["--logfile"],
+                        format='%(asctime)s - %(levelname)s - vmpoller-client[%(process)s]: %(message)s',
+                        level=logging.DEBUG)
+
     client = VMPollerClient(endpoint=args["--endpoint"], retries=args["--retries"], timeout=args["--timeout"])
 
     msg = { "type":    "hosts" if args["--hosts"] else "datastores",

src/vmpoller-proxy

@@ -30,6 +30,8 @@ between a pool of ZeroMQ workers.
 
 """
 
+import logging
+
 from vmpoller.core import VMPollerProxy, VMPollerClient
 from docopt import docopt
 
@@ -78,7 +80,7 @@ def status(endpoint):
 def main():
 
     usage="""
-Usage: vmpoller-proxy [-d] [-p <pidfile>] [-f <config-file>] start
+Usage: vmpoller-proxy [-d] [-p <pidfile>] [-f <config-file>] [-l <logfile>] start
        vmpoller-proxy -e <endpoint> stop
        vmpoller-proxy -e <endpoint> status
        vmpoller-proxy --help
@@ -99,10 +101,16 @@ Options:
   -f <config-file>, --config <config-file>  Specify config file to use
                                             [default: /etc/vmpoller/vmpoller-proxy.conf]
   -e <endpoint>, --endpoint <endpoint>      Specify the endpoint we connect to
+  -l <logfile>, --logfile <logfile>         Specify the logfile to use
+                                            [default: /var/log/vmpoller/vmpoller-proxy.log]
 
 """
 
     args = docopt(usage, version="1.0.0")
+
+    logging.basicConfig(filename=args["--logfile"],
+                        format='%(asctime)s - %(levelname)s - vmpoller-proxy[%(process)s]: %(message)s',
+                        level=logging.DEBUG)
 
     if args["start"]:
         start(args["--pidfile"], args["--config"], args["--daemon"])

src/vmpoller-worker

@@ -33,6 +33,8 @@ performing the actual polling from a vCenter server.
 
 """
 
+import logging
+
 from vmpoller.core import VMPollerWorker, VMPollerClient
 from docopt import docopt
 
@@ -80,7 +82,7 @@ def status(endpoint):
 
 def main():
     usage="""
-Usage: vmpoller-worker [-d] [-p <pidfile>] [-f <config-file>] start
+Usage: vmpoller-worker [-d] [-p <pidfile>] [-f <config-file>] [-l <logfile>] start
        vmpoller-worker -e <endpoint> stop
        vmpoller-worker -e <endpoint> status
        vmpoller-worker --help
@@ -101,10 +103,16 @@ Options:
   -f <config-file>, --config <config-file>  Specify config file to use
                                             [default: /etc/vmpoller/vmpoller-worker.conf]
   -e <endpoint>, --endpoint <endpoint>      Specify the endpoint we connect to
+  -l <logfile>, --logfile <logfile>         Specify the logfile to use
+                                            [default: /var/log/vmpoller/vmpoller-worker.log]
 
 """
 
     args = docopt(usage, version="1.0.0")
+
+    logging.basicConfig(filename=args["--logfile"],
+                        format='%(asctime)s - %(levelname)s - vmpoller-worker[%(process)s]: %(message)s',
+                        level=logging.DEBUG)
 
     if args["start"]:
         start(args["--pidfile"], args["--config"], args["--daemon"])

src/vmpoller/core.py

@@ -92,7 +92,7 @@
 import glob
 import json
 import time
-import syslog
+import logging
 import threading
 import ConfigParser
 from time import sleep
@@ -142,7 +142,7 @@ def run(self, config_file, start_agents=True):
 
         """
         if not os.path.exists(config_file):
-            syslog.syslog("Configuration file does not exists: %s" % config_file)
+            logging.error("Configuration file does not exists: %s", config_file)
             raise VMPollerException, "Configuration file does not exists: %s" % config_file
 
         config = ConfigParser.ConfigParser()
@@ -154,7 +154,7 @@ def run(self, config_file, start_agents=True):
             self.vcenter_configs = config.get('Default', 'vcenters')
             self.threads_num     = int(config.get('Default', 'threads'))
         except ConfigParser.NoOptionError as e:
-            syslog.syslog("Configuration issues detected in %s: %s" % (config_file, e))
+            logging.error("Configuration issues detected in %s: %s" , config_file, e)
             raise
 
         # A flag to signal that our threads and daemon should be terminated
@@ -185,17 +185,17 @@ def run(self, config_file, start_agents=True):
         try:
             self.mgmt.bind(self.mgmt_endpoint)
         except zmq.ZMQError as e:
-            syslog.syslog("Cannot bind management socket: %s" % e)
+            logging.error("Cannot bind management socket: %s", e)
             raise VMPollerException, "Cannot bind management socket: %s" % e
 
         # Create a ROUTER socket for forwarding client requests to our worker threads
-        syslog.syslog("Connecting to the VMPoller Proxy server")
+        logging.info("Connecting to the VMPoller Proxy server")
         self.router = self.zcontext.socket(zmq.ROUTER)
 
         try:
             self.router.connect(self.proxy_endpoint)
         except zmq.ZMQError as e:
-            syslog.syslog("Cannot connect worker to proxy: %s" % e)
+            logging.error("Cannot connect worker to proxy: %s", e)
             raise VMPollerException, "Cannot connect worker to proxy: %s" % e
 
         # Create a DEALER socket for passing messages from our worker threads back to the clients
@@ -297,15 +297,15 @@ def get_vcenter_configs(self, config_dir):
             
         """
         if not os.path.exists(config_dir) or not os.path.isdir(config_dir):
-            syslog.syslog("%s does not exists or is not a directory" % config_dir)
+            logging.error("%s does not exists or is not a directory", config_dir)
             raise VMPollerException, "%s does not exists or is not a directory" % config_dir
 
         # Get all *.conf files for the different vCenters
         path = os.path.join(config_dir, "*.conf")
         confFiles = glob.glob(path)
 
         if not confFiles:
-            syslog.syslog("No vCenter config files found in %s" % config_dir)
+            logging.error("No vCenter config files found in %s", config_dir)
             raise VMPollerException, "No vCenter config files found in %s" % config_dir
 
         return confFiles
@@ -319,7 +319,7 @@ def start_agents(self):
             try:
                 self.agents[eachAgent].connect()
             except Exception as e:
-                print 'Cannot connect to %s: %s' % (eachAgent, e)
+                logging.error('Cannot connect to %s: %s', eachAgent, e)
 
     def shutdown_agents(self):
         """
@@ -376,7 +376,7 @@ def process_mgmt_message(self, msg):
 
         if msg["cmd"] == "shutdown":
             self.time_to_die = True
-            syslog.syslog("VMPoller Worker is shutting down")
+            logging.info("VMPoller Worker is shutting down")
             return "Shutting down VMPoller Worker"
         elif msg["cmd"] == "status":
             header = "VMPoller Worker"
@@ -454,15 +454,15 @@ def get_host_property(self, msg):
             'runtime.bootTime':                              lambda p: time.strftime('%Y-%m-%d %H:%M:%S', p),
         }
 
-        syslog.syslog('[%s] Retrieving %s for host %s' % (self.vcenter, msg['property'], msg['name']))
+        logging.info('[%s] Retrieving %s for host %s', self.vcenter, msg['property'], msg['name'])
 
         # Get the properties for all registered hosts
         try:
             results = self.viserver._retrieve_properties_traversal(property_names=property_names,
                                                                    obj_type=MORTypes.HostSystem)
         except Exception as e:
             sleep(1) # Settle down for a moment
-	    syslog.syslog("Cannot get property for host %s: %s" % (msg["name"], e))
+	    logging.warning("Cannot get property for host %s: %s", msg["name"], e)
             return "Cannot get property for host %s: %s" % (msg["name"], e)
 
         # Do we have something to return?
@@ -545,14 +545,14 @@ def get_datastore_property(self, msg):
             property_names.extend(custom_zbx_properties[msg["property"]])
             property_names.remove(msg["property"])
 
-        syslog.syslog('[%s] Retrieving %s for datastore %s' % (self.vcenter, msg['property'], msg['name']))
+        logging.info('[%s] Retrieving %s for datastore %s', self.vcenter, msg['property'], msg['name'])
 
         try:
             results = self.viserver._retrieve_properties_traversal(property_names=property_names,
                                                                    obj_type=MORTypes.Datastore)
         except Exception as e:
             sleep(1) # Settle down for a moment
-            syslog.syslog("Cannot get property for datastore %s: %s" % (msg["name"], e))
+            logging.warning("Cannot get property for datastore %s: %s" % (msg["name"], e))
             return "Cannot get property for datastore %s: %s" % (msg["name"], e)
 
         if not results:
@@ -602,15 +602,15 @@ def discover_hosts(self):
         # Property <name>-<macros> mappings that Zabbix uses
         property_macros = { 'name': '{#ESX_NAME}', 'runtime.powerState': '{#ESX_POWERSTATE}' }
 
-        syslog.syslog('[%s] Discovering ESX hosts' % self.vcenter)
+        logging.info('[%s] Discovering ESX hosts', self.vcenter)
 
         # Retrieve the data
 	try:
             results = self.viserver._retrieve_properties_traversal(property_names=property_names,
                                                                    obj_type=MORTypes.HostSystem)
 	except Exception as e:
             sleep(1) # Settle down for a moment
-            syslog.syslog("Cannot discover hosts: %s" % e)
+            logging.warning("Cannot discover hosts: %s", e)
             return "Cannot discover hosts: %s" % e
 
         # Iterate over the results and prepare the JSON object
@@ -655,15 +655,15 @@ def discover_datastores(self):
                            'summary.accessible': '{#DS_ACCESSIBLE}',
                            }
 
-        syslog.syslog('[%s] Discovering datastores' % self.vcenter)
+        logging.info('[%s] Discovering datastores', self.vcenter)
 
         # Retrieve the data
 	try:
             results = self.viserver._retrieve_properties_traversal(property_names=property_names,
                                                                    obj_type=MORTypes.Datastore)
 	except Exception as e:
             sleep(1) # Settle down for a moment
-            syslog.syslog("Cannot discover datastores: %s" % e)
+            logging.warning("Cannot discover datastores: %s", e)
             return "Cannot discover datastores: %s" % e
 
         # Iterate over the results and prepare the JSON object
@@ -698,7 +698,7 @@ class VMPollerProxy(Daemon):
     """
     def run(self, config_file):
         if not os.path.exists(config_file):
-            syslog.syslog("Cannot read configuration for proxy: %s" % config_file)
+            logging.error("Cannot read configuration for proxy: %s", config_file)
             raise VMPollerException, "Cannot read configuration for proxy: %s" % config_file 
 
         config = ConfigParser.ConfigParser()
@@ -709,7 +709,7 @@ def run(self, config_file):
             self.backend_endpoint  = config.get('Default', 'backend')
             self.mgmt_endpoint     = config.get('Default', 'mgmt')
         except ConfigParser.NoOptionError as e:
-            syslog.syslog("Configuration issues detected in %s: %s" % (config_file, e))
+            logging.error("Configuration issues detected in %s: %s", config_file, e)
             raise
 
         # A flag to indicate that the VMPollerProxy daemon should be terminated
@@ -724,7 +724,7 @@ def run(self, config_file):
         try:
             self.mgmt.bind(self.mgmt_endpoint)
         except zmq.ZMQError as e:
-            syslog.syslog("Cannot bind management socket: %s" % e)
+            logging.error("Cannot bind management socket: %s", e)
             raise VMPollerException, "Cannot bind management socket: %s" % e
 
         # Socket facing clients
@@ -733,7 +733,7 @@ def run(self, config_file):
         try:
             self.frontend.bind(self.frontend_endpoint)
         except zmq.ZMQError as e:
-            syslog.syslog("Cannot bind frontend socket: %s" % e)
+            logging.error("Cannot bind frontend socket: %s", e)
             raise VMPollerException, "Cannot bind frontend socket: %s" % e
 
         # Socket facing workers
@@ -742,7 +742,7 @@ def run(self, config_file):
         try:
             self.backend.bind(self.backend_endpoint)
         except zmq.ZMQError as e:
-            syslog.syslog("Cannot bind backend socket: %s" % e)
+            logging.error("Cannot bind backend socket: %s", e)
             raise VMPollerException, "Cannot bind backend socket: %s" % e
 
         # Create a poll set for our sockets
@@ -751,7 +751,7 @@ def run(self, config_file):
         self.zpoller.register(self.backend, zmq.POLLIN)
         self.zpoller.register(self.mgmt, zmq.POLLIN)
 
-        syslog.syslog("Starting the VMPoller Proxy")
+        logging.info("Starting the VMPoller Proxy")
 
         # Enter the daemon loop from here
         while not self.time_to_die:
@@ -802,7 +802,7 @@ def process_mgmt_message(self, msg):
 
         if msg["cmd"] == "shutdown":
             self.time_to_die = True
-            syslog.syslog("VMPoller Proxy is shutting down")
+            logging.info("VMPoller Proxy is shutting down")
             return "Shutting down VMPoller Proxy"
         elif msg["cmd"] == "status":
             header = "VMPoller Proxy"
@@ -872,7 +872,7 @@ def run(self, msg):
             else:
                 # We didn't get a reply back from the server, let's retry
                 self.retries -= 1
-                syslog.syslog("Did not receive reply from server, retrying...")
+                logging.warning("Did not receive reply from server, retrying...")
 
                 # Socket is confused. Close and remove it.
                 self.zclient.close()
@@ -891,7 +891,7 @@ def run(self, msg):
 
         # Did we have any result reply at all?
         if not result:
-            syslog.syslog("Did not receive a reply from the server, aborting...")
+            logging.error("Did not receive a reply from the server, aborting...")
             return "Did not receive reply from the server, aborting..."
 
         return result