httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From r..@covalent.net
Subject Re: [PATCH] bringing down the server from an MPM thread
Date Wed, 26 Jul 2000 19:11:33 GMT

> > > If ap_poll() returns something other than APR_EINTR or APR_SUCCESS,
> > > with mpmt_pthread the child process goes away and the parent
> > > process creates a new child process to take its place.  ap_poll()
> > > again fails with the same error (as it will forever) for whichever
> > > process gets the accept mutex and so the process exits and a new one
> > > is created...
> > > 
> > > With prefork and a single listening socket, the error would be noticed
> > > on ap_accept().
> > 
> > But how is it handled with multiple listening sockets?  This is the case
> > that will most closely mimic what we are looking at with
> > mpmt_pthread.
> 
> See the previous paragraph ("If ap_poll().*with
> mpmt_pthread.*created...") for text on what happens with mpmt_pthread
> -- EIO from poll().

That isn't the question.  The question is what how does Apache 1.3 deal
with this and multiple listen sockets?  I asked about prefork, because
prefork is Apache 1.3 basically.

Ryan

_______________________________________________________________________________
Ryan Bloom                        	rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------


Mime
View raw message