Gaack. 8^( My problem here turns out to be the OS X network proxy layer, not the browser itself (the proxy layer that gets set in Network Preferences for a given link). It looks kind of hosed with persistent HTTP connections (or in the way we interact with it, or the way I set it up). I'll look more into this after I get through the basic testing and fixups.

On a better note, things look pretty good on FreeBSD 4.x and 2.2.x Linux. I want to do some throughput numbers next, after I finish some stress testing.

Chuck

On Friday, April 13, 2001, at 05:26 PM, Graham Leggett wrote:

Chuck Murcko wrote:

I see clusters of these, like from every server connected to the browser
in the (prefork) pool at once, and the browser is breaking the
connection(s).

Hmmm - if it's breaking all the connections in all the different processes at the same time it looks like it's a browser thing. Is it possible to test it with something neutral like ab? Which browser is this?

I have fixed all the error statements - they should all return APR
errors now. Can you try get another log file with the new error codes in
it?


Chuck Murcko Topsail Group http://www.topsail.org/

Reply via email to