httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dirk.vanGulik" <Dirk.vanGu...@jrc.it>
Subject Re: bug in 1.1 related to "unable to fork" problems?
Date Thu, 31 Oct 1996 09:54:56 GMT
> From owner-new-httpd@hyperreal.com Thu Oct 31 10:47:17 1996
> Subject: Re: bug in 1.1 related to "unable to fork" problems?
> To: new-httpd@hyperreal.com
> Date: Thu, 31 Oct 1996 07:40:57 +0000 (GMT)
> From: Ben Laurie <ben@gonzo.ben.algroup.co.uk>
> 
> Brian Behlendorf wrote:
> > 
> > On Wed, 30 Oct 1996, Ben Laurie wrote:
> > > I don't see how you can solve it. If you can't fork, you can't fork. Surely
> > > the error log shows this?
> > 
> > The error log does indeed show it.  I was just wondering if there was anything
> > which could be done about it.... at least, the current behavior is rather
> > suboptimal (one apache server spinning as fast as CPU allows, not answering
> > connections).  The ideal behavior would be one where the parent stays alive and
> > functional and continues to try forking children once every 10 seconds until it
> > succeeds, all the while bitching into the error_log.
> 
> Yeah, it shouldn't spin, this is true. That is definitely a bug.

 Just put in that sleep(10) seconds patch, posted some three-four weeks ago; we
have it in here; and it has surely solved our problems.

Dw.



Mime
View raw message