• Skip to main content

Uly.me

cloud engineer

  • Home
  • About
  • Archives

Parsing Wowza Logs

March 19, 2019

I’ve been wanting to look at Wowza logs for a very long time. I just spent a couple of hours writing a Bash script that generates a stats report. It parses the Wowza log file for the IP addresses, and runs them and against a site called ipinfo.io to get the viewer’s geo location. The output is filtered, sorted, takes a count, and finally generates a webpage. Here’s the Bash script.

# Can be placed in cron to run.
create-wowza-report.sh wowzastreamingengine_access.log

# Can be placed in cron to run. create-wowza-report.sh wowzastreamingengine_access.log

#!/bin/bash
 
# check for missing argument
if [ -z "$1" ]
  then
    echo "No arguments supplied. Format: create-wowza-report.sh wowzastreamingengine_access.log"
    exit 1
fi
 
# get the ips from the wowza log file
awk '{print $17}' $1 > output1.txt
 
# exclude invalid ips 
grep '[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}' output1.txt > output2.txt
 
# sort unique ips
cat output2.txt | sort -u | uniq > ips.txt
 
# pause for 1s
sleep 1s
 
# empty file just in case
> output3.txt
 
# set filename
filename="ips.txt"
 
# read filename and get geo info
while read -r ipaddress; do
  curl ipinfo.io/"$ipaddress" 2>/dev/null | awk -F'"' '$2=="city"{printf("%s, ", $4)}$2=="region"{print $4}' >> output3.txt
done < "$filename"
 
# cleanup empty geo locations
grep -v '^,' output3.txt > output4.txt
 
# sort output alphabetically
cat output4.txt | sort > output5.txt
 
# get a count of connections
count=$(wc -l < output5.txt)
 
# get today's date
filename=$(date +%Y%m%d)
date=`date +'%A %B %d, %Y'`
 
# rename file
mv output5.txt $filename.txt
 
# generate web page
cat <<EOF > $filename.php
<html>
<head>
<title>Wowza Streaming Stats on $date</title>
</head>
<body>
<h1>Live Streaming Stats</h1>
<p>Number of connections on $date: $count</p>
<?php include('$filename.txt');?>
</body>
</html>
EOF
 
# move files to web directory
cp $filename.* /var/www/html
 
# cleanup files
rm output*.txt
rm ips.txt

#!/bin/bash # check for missing argument if [ -z "$1" ] then echo "No arguments supplied. Format: create-wowza-report.sh wowzastreamingengine_access.log" exit 1 fi # get the ips from the wowza log file awk '{print $17}' $1 > output1.txt # exclude invalid ips grep '[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}' output1.txt > output2.txt # sort unique ips cat output2.txt | sort -u | uniq > ips.txt # pause for 1s sleep 1s # empty file just in case > output3.txt # set filename filename="ips.txt" # read filename and get geo info while read -r ipaddress; do curl ipinfo.io/"$ipaddress" 2>/dev/null | awk -F'"' '$2=="city"{printf("%s, ", $4)}$2=="region"{print $4}' >> output3.txt done < "$filename" # cleanup empty geo locations grep -v '^,' output3.txt > output4.txt # sort output alphabetically cat output4.txt | sort > output5.txt # get a count of connections count=$(wc -l < output5.txt) # get today's date filename=$(date +%Y%m%d) date=`date +'%A %B %d, %Y'` # rename file mv output5.txt $filename.txt # generate web page cat <<EOF > $filename.php <html> <head> <title>Wowza Streaming Stats on $date</title> </head> <body> <h1>Live Streaming Stats</h1> <p>Number of connections on $date: $count</p> <?php include('$filename.txt');?> </body> </html> EOF # move files to web directory cp $filename.* /var/www/html # cleanup files rm output*.txt rm ips.txt

Filed Under: Linux Tagged With: bash, live streaming, logs, stats, wowza

Search This Website

Subscribe Via Email

  • Home
  • About
  • Archives

Copyright © 2023