Unicorn Timeout

Question

I am wondering what is the best way to track unicorn timeouts from my application.

There are a few bits of the application that are slow, and they currently time out quietly. I could increase the timeout, but this makes the problem under the carpet difficult. Ideally, I would like to receive an air brake notice or something like that. I am not looking for guidance on efficiency, just a way to effectively and reliably report timeouts.

How do other people handle timeouts from their rails application using a unicorn? scan nginx error log? to connect a handler in a unicorn configuration? handler in nginx configuration?

[Ubuntu 12.04 + nginx + unicorn + rails 3.2 + ruby ​​1.9.3]

Answer 1

I would be inclined to create an account in New Relic , use the newrelic_rpm and collect performance data related to this problem in their web application. NR offers a free account level that is useful for this situation.

The advantage of using this somewhat lazy instrumental method is twofold: there is no requirement for implementing custom measuring equipment in your application; and there is no need to check your logs in a downstream web server for 503 server errors.

Answer 2

Refer to the 'configurator section of the Unicorn manual website. There are a number of points around lag and timeouts that should give you the control you need.