From owner-freebsd-hackers Tue Oct 27 22:43:55 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA25512 for freebsd-hackers-outgoing; Tue, 27 Oct 1998 22:43:55 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA25505 for ; Tue, 27 Oct 1998 22:43:53 -0800 (PST) (envelope-from marcs@znep.com) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.9.1a/8.9.1) with UUCP id XAA03972; Tue, 27 Oct 1998 23:42:59 -0700 (MST) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with ESMTP id WAA04817; Tue, 27 Oct 1998 22:43:47 -0800 (PST) Date: Tue, 27 Oct 1998 22:43:47 -0800 (PST) From: Marc Slemko To: Mike Smith cc: Steven Yang , "'freebsd-hackers@freebsd.org'" Subject: Re: FW: Can't get rid of my mbufs. In-Reply-To: <199810280635.WAA00967@dingo.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, 27 Oct 1998, Mike Smith wrote: > > SUre, the FastCGI stuff is probably doing something more complex than > > static files and it does double the total transfered (from fastcgi to the > > server, from the server to the client), but still... > > Hmm, I wasn't aware that the fastcgi reference implied a second server. > How does it communicate with the server proper? INET domain > connection(s)? Local domain connections? Named pipes? Shared memory? TCP connection. Well, ok, I guess that isn't true. It can use a pipe and probably would for local connections. OTOH, I have no idea what version is being used or how it is configured, and I'm no FastCGI expert anyway. > It sounds like there might be room there for many mbufs to be > legitimately consumed, effectively leaked by the application > interaction. Possibly. I would recommend simplifying the test case to just Apache serving a 20k static file and see if that changes things. And, of course, the netstat -n output that dg asked for to see if there is anything legit. If it hasn't been tried already, killing all the Apache and fastcgi processes and anything related and seeing if use drops down would be another possible idea. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message