- basics of init-script working

- basic shell script which enables virtualenv environment and calls traceroute.py at intervals
- added stubs to only call webhook based on a specified latency
- still need to store offline data in redis - coming soon

README.md

@@ -7,6 +7,7 @@ Multi-source traceroute with geolocation information. Demo: [IP Address Lookup](
 
 1. Install dependencies listed in requirements.txt file. Pip is recommended. You could also use virtualenv to keep things isolated.
 2. Save traceroute.py into a directory with its path stored in your PYTHONPATH environment variable. (if using virtualenv, copy it here)
+3. Install Redis-Server ( may change )
 
 
 ## Usage

init-script/traceroute

@@ -0,0 +1,64 @@
+#!/bin/bash
+# traceroute.py daemon
+# chkconfig: 345 20 80
+# description: traceroute.py daemon
+# processname: traceroute.py
+
+DAEMON_PATH="/var/lib/python/traceroute"
+
+DAEMON=traceroute.sh
+DAEMONOPTS=""
+
+NAME=traceroute.py
+DESC="Traceroute Python Tool"
+PIDFILE=/var/run/$NAME.pid
+SCRIPTNAME=/etc/init.d/$NAME
+
+case "$1" in
+start)
+	printf "%-50s" "Starting $NAME..."
+	cd $DAEMON_PATH
+	PID=`$DAEMON $DAEMONOPTS > /dev/null 2>&1 & echo $!`
+	#echo "Saving PID" $PID " to " $PIDFILE
+        if [ -z $PID ]; then
+            printf "%s\n" "Fail"
+        else
+            echo $PID > $PIDFILE
+            printf "%s\n" "Ok"
+        fi
+;;
+status)
+        printf "%-50s" "Checking $NAME..."
+        if [ -f $PIDFILE ]; then
+            PID=`cat $PIDFILE`
+            if [ -z "`ps axf | grep ${PID} | grep -v grep`" ]; then
+                printf "%s\n" "Process dead but pidfile exists"
+            else
+                echo "Running"
+            fi
+        else
+            printf "%s\n" "Service not running"
+        fi
+;;
+stop)
+        printf "%-50s" "Stopping $NAME"
+            PID=`cat $PIDFILE`
+            cd $DAEMON_PATH
+        if [ -f $PIDFILE ]; then
+            kill -HUP $PID
+            printf "%s\n" "Ok"
+            rm -f $PIDFILE
+        else
+            printf "%s\n" "pidfile not found"
+        fi
+;;
+
+restart)
+  	$0 stop
+  	$0 start
+;;
+
+*)
+        echo "Usage: $0 {status|start|stop|restart}"
+        exit 1
+esac

init-script/traceroute.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+
+EXEC_DIR=/var/lib/python/traceroute
+VIRTUALENV_DIR=/var/lib/python/traceroute/env
+
+# Repeat every 5 seconds
+INTERVAL=5 #
+
+source $VIRTUALENV_DIR/bin/activate
+
+cd $EXEC_DIR
+
+# TODO add some logging later
+
+while true;
+do
+    python ./traceroute.py --ip_address=8.8.8.8 -c LO --webhook=http://localhost:8081/test;
+    sleep $INTERVAL;
+done

traceroute.py

@@ -27,7 +27,7 @@ class Traceroute(object):
     Multi-source traceroute instance.
     """
     def __init__(self, ip_address, source=None, country="US", tmp_dir="/tmp",
-                 no_geo=False, timeout=120, debug=False):
+                 no_geo=False, timeout=120, debug=False, max_latency=5):
         super(Traceroute, self).__init__()
         self.ip_address = ip_address
         self.source = source
@@ -37,6 +37,9 @@ class Traceroute(object):
             self.source = sources[country]
         self.tmp_dir = tmp_dir
 
+        self.LATENCY_THRESHOLD = float(max_latency)
+    
+
         self.no_geo = no_geo
         self.timeout = timeout
         self.debug = debug
@@ -44,6 +47,9 @@ class Traceroute(object):
         self.hops = {}
         self.country = country
 
+        # flag to determine if webhook alert is warranted
+        self.latency_exceeded = False
+
         # Localhost Specific operations happen here
         if self.country == 'LO':
             self.local_mode = True
@@ -59,6 +65,11 @@ class Traceroute(object):
         self.__run_traceroute() 
         self.probe_end   = time.time() * 1000
 
+    def pingLatencyThresholdExceeded(self):
+        """public method to query state of Traceroute calls"""
+
+        return self.latency_exceeded
+
     def __run_traceroute(self):
         """
         Instead of running the actual traceroute command, we will fetch
@@ -134,6 +145,7 @@ class Traceroute(object):
         hop_element_pattern = '([\d\w.-]+)\s+\((\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})\)\s+(\d+\.\d+ ms)'
         hp  = re.compile(hop_element_pattern)
 
+        alertTriggered = False
         for entry in traceroute.split('\n'):
             entry = entry.strip()
             result = re.match(hop_pattern,entry)
@@ -146,9 +158,17 @@ class Traceroute(object):
             hop_hosts   = re.findall(host_pattern, hop['hosts'])
 
             self.hops[hop_num] = []
+            
             for host in hop_hosts:
                 m = hp.search(host)
                 (hostname, ip, ping_time) = m.groups()
+                
+                # Check ping time to see if it exceeds threshold. Once one is found, don't need any more info from other hops
+                if alertTriggered is False:
+                    if self._exceeds_hop_latency(ping_time):
+                        self.latency_exceeded   = True
+                        alertTriggered          = True
+
                 if self.no_geo:
                     self.hops[hop_num].append(
                         { 
@@ -207,6 +227,15 @@ class Traceroute(object):
             if 'latitude' in tmp_location and 'longitude' in tmp_location:
                 location = tmp_location
         return location
+    def _exceeds_hop_latency(self,ping_time):
+        """return true if hop time exceeds specified latency threshold"""
+        # remote ' ms' from ping time
+        ping_as_float = float(ping_time.replace(" ms",""))
+        return ping_as_float >= self.LATENCY_THRESHOLD
+
+
+
+
 
     def execute_cmd(self, cmd):
         """
@@ -360,6 +389,18 @@ def post_result(webhook_url, report, timeout=120):
     """
     return requests.post(webhook_url, data=json.dumps(report), timeout=timeout)
 
+def webhook_available(webhook_url):
+    """ 
+    Function to check if a webhook host is responding.
+    Not 100% sure this will work...
+    """
+    try:
+        data = urllib.urlopen(webhook_url)
+        return True
+    except Exception,e:
+        return False
+
+
 
 def main():
     cmdparser = optparse.OptionParser("%prog --ip_address=IP_ADDRESS")
@@ -393,6 +434,11 @@ def main():
     cmdparser.add_option(
         "-w", "--webhook", type="string", default="",
         help="Specify URL to POST report payload rather than stdout")
+    
+    cmdparser.add_option(
+        "--max_latency", type="int", default="5",
+        help="Maximum latency whereby the system will trigger the webhook ( if requested ). ")
+    
 
     options, _  = cmdparser.parse_args()
     json_file   = open(options.json_file, "r").read()
@@ -406,17 +452,32 @@ def main():
                             tmp_dir=options.tmp_dir,
                             no_geo=options.no_geo,
                             timeout=options.timeout,
-                            debug=options.debug)
+                            debug=options.debug, max_latency = options.max_latency)
 
     # pull complete report -> Hop data plus meta info about the network
     report = traceroute.get_report()
 
+
+ 
+
     if options.webhook != "":
-        try:
-            result = post_result(options.webhook, report, options.timeout)
-            print "Webhook POST Result: {}".format(result)
-        except Exception,e:
-            print "Provided webhook {0} is invalid. Message was: {1}".format(options.webhook, e)
+        # check if remote host is available
+        if webhook_available(options.webhook):
+
+            # if available, check if there are any outstanding reports that should be sent
+            # if traceroute::backlog == true => Purge results
+            
+            if traceroute.pingLatencyThresholdExceeded():
+                try:
+                    result = post_result(options.webhook, report, options.timeout)
+                    print "Webhook POST Result: {}".format(result)
+                except Exception,e:
+                    print "Provided webhook {0} is invalid. Message was: {1}".format(options.webhook, e)
+        else:
+            print "Webhook unavailable, caching"
+            pass
+            # Dump Result into Redis
+            # Set redis flag traceroute::backlog => true
     else:
         print(json.dumps(report, indent=4))
     return 0