Troubleshooting Linux monitoring

Follow

*THIS ARTICLE IS FOR FIXING THE MONITORING OF A SELF-MANAGED SERVICE*

Servers Australia offers free optional monitoring for all virtual and physical servers.

This includes preventative checks such as monitoring CPU load and free disk space. While other checks are for critical services like ssh, mail or http.

If the server has a server management plan with Servers Australia, then we'll take care of making sure all checks are working and up to date.

If you have elected to have your server self-managed, then we'll still alert you with what our monitoring picks up, but you will be required to fix the issue yourself.

If we find that your monitoring isn't working, we will alert you. We will request that you check the status of the monitoring so it can be corrected on your server.
We won't log into the server and correct it ourselves

Failure to correct the broken monitoring after being requested will result in the server being removed from our monitoring program. We will then have no visibility of your server and it will be completely up to you to manage the server.

 

Here are the most problems and fixes to the monitoring:

1. "Connection Refused"

This problem occurs when our monitoring server gets refused when attempting to connect to the server to read it's status. 

How to fix it:

Check that the firewall isn't blocking access to and from 202.130.35.150 and that port 5666 is open for it.
Also make sure there is a line in /etc/nagios/nrpe.cfg with
allowed_hosts = 202.130.35.150

2. "Connection timed out"

This means that the server wasn't refused, but it also didn't get any data from your server

How to fix it:

Make sure that the nrpe service is running. You can restart the service using service nrpe restart

 

3. "Unable to read output"

This occurs when the data retrieved from your server is incomprehensible to our monitoring.

How to fix it:

Check that something something ssl something

 

Have more questions? Submit a request

Comments

Powered by Zendesk