Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am trying to load test an amazon centos instance sitting behind the amazon loadbalancer.

The application is written on php and uses the elastic cache and media is stored in s3 bucket.

When the server gets more than 1000 concurrent requests its starts giving "Apache interanl dummy connection" and the pages dont load when tried externally, even though the load uptime metric does not go beyond 1.

Which means theres plenty of CPU and MEM resources, I have configured the apache prefork to this:

StartServers       32
MinSpareServers    20
MaxSpareServers   80
ServerLimit      256
MaxClients       256
MaxRequestsPerChild  4000

but because the resources are not used so this would not make any effect, so I think i need to configure apache to handle php requests.

Has anyone had any scenario like this?

share|improve this question

For a start, increase ServerLimit and MaxClients to be more than 1000. That will allow 1000 concurrent connections. Depending on how your server is used by the clients, you'll want to increase MaxSpareServers and MinSpareServers a little too.

Just a warning, you'll start eating up a lot more memory with more apache threads running. Memory might become an issue.

share|improve this answer
ServerLimit shouldn't be blindly increased to 1k without knowing what you're doing. If you set it to something higher than your server can handle in available RAM, it's going to generally do more harm than good. – ceejayoz Dec 23 '11 at 20:41

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.