Monitoring your own server
We see the port, the power and the machine. We cannot see your application, because we are not in it. That half is yours.
The division
Our monitoring covers the network, the links, the routing and whether the hardware is alive. If a switch port goes down, we know before you do. If your database stops accepting connections while the machine happily answers pings, nothing on our side notices — and you would not want us watching inside your machine anyway.
What to watch, in order of usefulness
- Disk space. The number one cause of servers that stop working for no obvious reason. Alert at 80 percent, not at 99.
- Does the service answer? Not “is the port open” but a real request that returns a real answer.
- Certificate expiry. Thirty days ahead, by mail. Renewals break silently.
- Load and memory — see when the server runs out of memory.
- Disk health — see SMART. Tell us as soon as the numbers move.
- Backups. Alert when a backup did not run. Silence is not success.
Watch from outside as well
Monitoring that runs on the machine it monitors cannot tell you the machine is gone. Put one check somewhere else — another provider, a free uptime service, a script on your laptop — that fetches a real page and shouts if it does not get one.
Keep the alerts few
An alert you ignore is worse than no alert, because it teaches you to ignore the next one. Five checks you act on beat fifty you filter into a folder.
Still stuck? Mail support@novogara.com — an engineer answers, at any hour. Back to the knowledge base