Quick and dirty webserver monitoring script.

[KuJoe]

Here's a simple little script I use on my webservers, very helpful to get your website back online when you're not around.

http=$(curl --write-out %{http_code} --silent --output /dev/null http://localhost)
if [ $http -ne 200 ] ; then
        echo "nginx is down at `date`" >> /var/log/httpchk.log
        /etc/init.d/nginx restart
else
        echo "nginx is up at `date`" >> /var/log/httpchk.log
fi

Just replace nginx with your webserver of choice. For MySQL powered sites I usually have it restart MySQL also in case that is the cause of the site being down.

 

All you need to do is add this code to a new file (I use /root/httpchk.sh) and then add the following to your crontab file (crontab -e on most servers):

*/5 * * * * /bin/sh /root/httpchk.sh

I find this comes in handy when you are using a VPS with small amounts of RAM or during various small attacks that fill up your webserver's connections.

 

I had somebody ask about adding code for sending an e-mail, I considered doing this but I have 2 monitoring services (1 custom and NodePing) that alert me of any outage and provide me the error code but I will expand on this code with that shortly.

 

Thanks Jono20201 for filling in some blanks I left out.

 

[Fritz]

Sorry for being noob here. where should I put that?

 

[Jono20201]

Sorry for being noob here. where should I put that?

Paste it in a shell file somewhere on your Linux system (for instance called check.sh in /root/) and then create a cron to check it (crontab -e):

*/5 * * * * /bin/sh /root/check.sh

 

[drmike]

Nice light script @KuJoe!

 

[Hassan]

Here's a simple little script I use on my webservers, very helpful to get your website back online when you're not around.



http=$(curl --write-out %{http_code} --silent --output /dev/null http://localhost)
if [ $http -ne 200 ] ; then
echo "nginx is down at `date`" >> /var/log/httpchk.log
/etc/init.d/nginx restart
else
echo "nginx is up at `date`" >> /var/log/httpchk.log
fi

Just replace nginx with your webserver of choice. For MySQL powered sites I usually have it restart MySQL also in case that is the cause of the site being down. I find this comes in handy when you are using a VPS with small amounts of RAM or during various small attacks that fill up your webserver's connections.

Going to use this on some of my managed clients VPS', thanks KuJoe

 

[KuJoe]

Paste it in a shell file somewhere on your Linux system (for instance called check.sh in /root/) and then create a cron to check it (crontab -e):


*/5 * * * * /bin/sh /root/check.sh

Thanks, I should have been more clear in my original post but I was a bit tired. I'll append your post to mine for clarity.

 

[BlackoutIsHere]

Thanks, this clearly follows KISS and I like it.