Den 07/10/10 20.47, Nathan Andelin skrev:
From: Thorbjoern Ravn Andersen
If I understand Aaron correctly, the problem is the HTTP server not
servicing requests fast enough, not necessarily that there are too few
threads. The behaviour described fits with the TCP/IP layer filling up.
I subscribed the behavior to the HTTP server rejecting connections, probably due
to limiting the number of threads, and at the same time enabling persistent
connections. If you only enable 10 connections, and hold on to each, somebody
is bound to get rejected.

I've done quite a bit of stress testing, and never encountered a rejected
connection at the i/OS level; only at the HTTP server level.
I believe the TCP/IP layer refuses connections if the connections are not acknowledged fast enough by the listening program. In your situation there should be enough threads that the HTTP-server does not deny connections by it self and the load should be high enough that the HTTP-server cannot process the requests fast enough.

What loads (originating systems, simultaneous users, request intervals) were you testing with?


As an Amazon Associate we earn from qualifying purchases.

This thread ...

Follow-Ups:
Replies:

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].

Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.