Re: StartCGI bug?

I'm not sure if concurrent requests are made asynchronously or not.

If logging is configured for your HTTP server instance, you can check the
timestamp of each request to see whether multiple requests are made in say
< 1 second. The HTTP server automatically starts new CGI Jobs under load,
whenever a new one is needed.

StartCGI 5 is indicated and there are 2 pools of 5 CGI jobs each time.


By "2 pools", I assume you're referring to 2 HTTP Server Instances.

When the server is started
there is 1 pool of 5 jobs, then after the first request there are 2 pools
of 5.

That doesn't make sense to me.

why is the first request taking so long

I would guess some sort of resource contention during activation.

why do I have
twice as many jobs running afterwards as indicated in StartCGI?

Again, that doesn't make sense to me. You seem to be saying that there are
2 pools of CGI Jobs for a single HTTP Server instance. What does pool mean,
for you?