worker_connections are not enough

If you have a busy Accelerator you might see a set of errors like this in your log —

2014/01/25 09:06:52 alert 7345#0: 5000 worker_connections are not enough
2014/01/25 09:06:52 alert 7345#0: 5000 worker_connections are not enough
2014/01/25 09:06:52 alert 7345#0: 5000 worker_connections are not enough
2014/01/25 09:06:52 alert 7345#0: 5000 worker_connections are not enough

That means you are hitting the connection limit you set. If you are experiencing an attack this might be desired, but if not you need to increase the number of clients allowed.

Go to Accelerator and adjust your “Max Connections” and “Worker Processes” numbers. You should have one Worker Process per core you have, and the max connections is per worker process.

So if you have 8 cores and set it to 5000 you will allow 40,000 connections at once.

Was this article helpful?
0 out of 0 found this helpful

Comments

0 comments

Article is closed for comments.