I'm migrating my servers from Rackspace over to Amazon. Our servers are running mod_perl and the connect times are much longer on the Amazon server (oftentimes up to 1 second just to connect) with the same config. Ten concurrent connection can be 100-500ms in connect time alone.

The actually processing times are very fast. The perl script we're using for testing doesn't dip into a database.

Changing the size of the EC2 instance does not affect the connect time in any way.

I assume that means that something's up with the Apache config, but I don't know where to look.

Can anyone recommend where to look/troubleshoot?

  • Is the name lookup on into your apache config ? – silviud Apr 20 '13 at 17:20
  • Are you using mod_perl handlers, Registry, or CGI? MPM? Server/Clients configuration lines? It's normal for the first connection to take a while per server thread, unless you precompile your handlers in the startup script. – Chris S Apr 21 '13 at 3:43
  • Pertinent lines: HostnameLookups Off, LoadModule perl_module modules/mod_perl.so, ModPerl::Registry is off for testing, but will be needed in production. MaxClients 150, StartServers 5, MinSpareServers 5, MaxSpareServers 10, MaxRequestsPerChild 0, KeepAlive off. – Disco Apr 22 '13 at 17:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.