All of a sudden, apache has starting to go crazy.

It will not run more than 3 minutes at a time before it does not accept any more request.

I have tried accessing it from many different isp's, proxies, etc, all the same. Also, we have several monitoring services, so I know it's not my connection.

I can run ./Apache status and see that many pids are running even after it cuts out. After it cuts out, I restart it and everything is fine but then 3 minutes later it starts up again.

The only thing that I ran since it was normal and now was up2date for red hat linux but I doubt that is the case. Nothing else for the past few weeks.

Here are the httpd.conf variables but I have tried changing these based on suggestions I found on the web and nothing.

ServerType standalone
ServerRoot "/usr/local/apache"
PidFile /usr/local/apache/logs/httpd.pid
Timeout 300
KeepAlive On
MaxKeepAliveRequests 100
KeepAliveTimeout 15
MinSpareServers 5
MaxSpareServers 100
StartServers 5
MaxClients 150

Everything else is either off or disabled.

Again, I have tried changing these numbers every which way based on suggestions on apache.org and other sites but nothing works, so I went back to the defaults.

Any ideas?

Hi!

I have a similar problem, do you find any solutions?

Thanks!

Hi

What's the output from your error logs?

Are your server loaded?

Try doing a netstat to see the list of connections when the server stops responding. Pipe it into "wc -l" to get the number of connections.

You might be experiencing a too high number of connections (either big load or DOS attack). That could cause the exact type of behaviour you're describing (I've seen it before).

Try checking your kernel settings for the max number of connections set for ipv4.