On Mon, 4 Jul 2005 10:19:14 +0200
"Peter J. Holzer" <[EMAIL PROTECTED]> wrote:
> On 2005-07-03 19:11:16 -0700, Robert Spier wrote:
> > > Forkserver tends to die with segmentation faults in the signal
> > > handler as several people on the list noticed. It also sometimes
> > > misses the death of a child. This patch moves the REAPER into the
> > > main loop. It now checks for dead children after each accept and once
> > > every second if there are more then $MAXCONN children.
> >
> > It's working fine for a lot of folks without the patch.
Race conditions are bad news whether they trigger at low load or not.
It's usually the extreme cases that trigger them reliably, but you
can bet they'll bite you in the ankles now and then too.
To handle the infrequent connection and spiky connection cases, why
not use IO::Select on the server socket, so you can wake up every
config('max-hibernate') seconds and give your zombie armies their
last rites?
But then, as a high load user I ran out of both memory and cpu and had to
run off with high_perf. High_perf is stable for me now, BTW.
Brian
