monitor.py

from monascaclient import client as monclient, ksclient
import monascaclient.exc as exc
import socket
import time

LOG_PATH = "/path/to/log"
last_checked = ''

# A metric identity is composed by its name and dimensions (a dictionary
# that represents attributes, as the host of the metric owner and the 
# application id). For this example, socket.getfqdn() gets the context hostname.
metric_info = {'name': 'metric.name', 
	       'dimensions': {'hostname': socket.getfqdn()}}

# Monasca client needs authenticate via keystone. Creating a keystone 
# client is possible to get a token that will be used after to create a 
# Monasca client.
ks = ksclient.KSClient(
            auth_url='http://<keystone_url>:5000/v3',
            username='user',
            password='password',
            project_name='monasca'
        )
        
# Creating a monasca client using the keystone token
monasca_client = monclient.Client(
            '2_0', ks.monasca_url, token=ks.token
        )

# For this approach, the format of the log must follow some pattern.
# The method get_metric_value extract the value from this log pattern.
# Flags are used to capture logs of interest and value into these logs
# For this use case "#" is a token to identify where exactly the value 
# is located and makes possible to convert the string to float value.
def get_metric_value(log):
	
	for i in range(len(log)-1, 0, -1):
		if log[i] == '#':
			value = float(log[i+1:len(log)])
				
	return value

# This loop checks new logs continously and publish a new measurement into
# Monasca when it is identified.
while True:
        log = open(LOG_PATH, "r")
        latest_log = log.readlines()
	# Check if log file is still empty
	if latest_log != []:
		latest_log = latest_log[-1]
        
        # Check if this log line contains a new metric measurement
        if '[Progress]' in latest_log and last_checked != latest_log:
                last_checked = latest_log
                # Add to metric_info values for this measurement:
                # value and timestamp
                value = get_metric_value(latest_log)
                metric_name['name'] = 'progress'
                metric_info['value'] = value
                metric_info['timestamp'] = time.time() * 1000
                # Sending the metric to Monasca
                monasca_client.metrics.create(**metric_info)
                print "Progress published: %.2f" % (value)

        elif '[Elapsed Time]' in latest_log and last_checked != latest_log:
                last_checked = latest_log
                # Add to metric_info values for this measurement:
                # value and timestamp
                value = get_metric_value(latest_log)
                metric_info['name'] = 'elapsed_time'
                metric_info['value'] = value
                metric_info['timestamp'] = time.time() * 1000
                # Sending the metric to Monasca
                monasca_client.metrics.create(**metric_info)
                print "Elapsed Time published: %.2f" % (value)                
        
        # Flag that checks if the log capture is ended
        elif '[END]' in latest_log:
                break