Monitoring your web instance

Pre-condition

You have just launched your web instance but you can not make sure that if it would run appropriately within your launching timing.

You want to specify a reliable way to detect whether when your system is down.

AWS Monitoring Tools

In AWS management console, you can use CloudWatch management to create multiple alarms for detecting when instance fails to run

There are various kinds of metrics and alarms but we should focus on EC2 instances only.

For EC2 instances, I recommend you to create at least 3 kinds of alarms : High status CPU, High CPU usage and High Disk Write Volume.

Those alarms pretty much cover all cases that instance is down, so whenever instance are down or get an unexpected 50x error, you will get an alert email for it. Remember to set the interval for CloudWatch, the minimum interval time is set to 2 minutes.

Non-AWS Monitoring Tools

For non AWS instance basically cover all instances. Somes would have a dedicated service like AWS CloudWatch ( Azure for example ) but for some free services, we highly doubt it will have such services, so the only reliable way is to use some third party service.

For some third party service out there, we recommend Uptime Robot, It is totally free and offer us a 5-minute-interval checking status, and if our index have some unexpected error ( like 50x error or instance is down ) it will send you an email asap.

The drawback of this service is that it needs to specify the URL so for 1 monitor, you only can check for one specific route and normally it would be your index page, you can not specify some other get or post request for example
If you liked this article

Let's subscribe the updates of Scuti!
Share on Google Plus

About Pham Tuan Anh

This is a short description in the author block about the author. You edit it by entering text in the "Biographical Info" field in the user admin panel.
    Blogger Comment
    Facebook Comment

0 Comments:

Post a Comment