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 |
#!/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 |