From owner-freebsd-hackers Fri May 22 17:38:16 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA28326 for freebsd-hackers-outgoing; Fri, 22 May 1998 17:38:16 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from biggusdiskus.flyingfox.com (biggusdiskus.flyingfox.com [205.162.1.28]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA28321 for ; Fri, 22 May 1998 17:38:08 -0700 (PDT) (envelope-from jas@flyingfox.com) Received: (from jas@localhost) by biggusdiskus.flyingfox.com (8.8.5/8.8.5) id RAA04688; Fri, 22 May 1998 17:39:21 -0700 (PDT) Date: Fri, 22 May 1998 17:39:21 -0700 (PDT) From: Jim Shankland Message-Id: <199805230039.RAA04688@biggusdiskus.flyingfox.com> To: njs3@doc.ic.ac.uk, Studded@san.rr.com Subject: Re: tcp states and sysctl's Cc: hackers@FreeBSD.ORG Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Studded writes: [Re all those TCP connections in FIN_WAIT_2]: > My main concern is fd's right now (mostly due to a project I'm working > on :). When exactly does the fd associated with a connection like this > get allocated and when does it get released? The fd is gone; the server has closed the socket. In my opinion, the biggest (potential) problem with having thousands of (de facto dead) TCP connections hanging around in FIN_WAIT_2 is the fact that there are still some linear traversals of the TCB list in the code (like in the timers). My apologies in advance if this has been addressed in -CURRENT; I "live" on the STABLE branch. Jim Shankland Flying Fox Computer Systems, Inc. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message