Yes, and yes. I've been doing that, but nothing yet.
It does look like the memory problem is fixed; however, I still cannot access akamai border servers via the proxy. No more free() errors in the logs.
Connections terminate with the usual pipe write error; suggesting that the browser is giving up before us.
Tcpdump shows normal negotiation, then we fail to ack data. All (4) client connections' processes go into sbwait state on FreeBSD 4, same as any connection relaying data. But they get stuck there.
Prefork mpm. I see the same on Linux here. You're on Linux PPC, right? I should install that here, finally.
Chuck
On Friday, May 4, 2001, at 07:59 PM, Graham Leggett wrote:
Chuck Murcko wrote:
Actually the Apple site is the OS X site:
http://www.apple.com/macosx/
www.apple.com never had the problem.
That site works great here...
BTW, Graham, that's all I'm doing here with the proxy test setup, basically just the minimum proxy config with default everything.
My proxy isn't on localhost, though. It's a couple of hops away from the
client box.
The problem went away a few days ago after updating my sandbox to the latest versions of everything. I haven't made any changes to the proxy code apart from a fix to the PROXY_DIR filter that turns directory listings into HTML-ised pages. Currently it would seem the bug was elsewhere in Apache and was fixed. There have been a whole bunch of changes to the Apache core lately.
Are the systems the same as before? Is it possible to watch what is going on with a debugger?
Regards, Graham -- ----------------------------------------- [EMAIL PROTECTED] "There's a moon over Bourbon Street tonight..."
Chuck Murcko Topsail Group http://www.topsail.org/
