Monitoring Process

LPH

Well-known member
My sites were scattered among many different shared servers. Over the past year and a half they were put together onto one shared server. While the host was very polite, the amount of resources taken up was getting rude on my part. I decided to strike out on my own, buy a droplet from Digital Ocean, and start playing testing.

The 512 MB RAM was clearly too low and the database connection would drop after the test site was put up online. Using tail -f /var/log/messages was blank for me but support said this was the reason and so the droplet was upgraded to 1 GB.

Now I have top running and the D.O. graph shows the same spikes of 50% CPU usage and top shows bumps every few seconds of mysql to 11%.

This brings me to my question ...

What are you using to monitor your servers?
What would you recommend a newbie use?
When something goes wrong, do you have a procedure to track down the issue?

Also, since this server is just a default droplet, it feels like the site is hanging out there in the wild. What type of protection should be done? What type of procedure would you use to optimize the installation?
 
Some people use services like pingdom I believe, but if you have an extra PC at home you could set up nagios (or similar) and monitor anything on the machine.. server load, connectivity, etc.. for free.
 
  • Like
Reactions: LPH
Top Bottom