From owner-freebsd-current Sat Sep 5 21:31:38 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA16239 for freebsd-current-outgoing; Sat, 5 Sep 1998 21:31:38 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from nagual.pp.ru (lsd.relcom.eu.net [193.125.27.73]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA16222; Sat, 5 Sep 1998 21:31:33 -0700 (PDT) (envelope-from ache@nagual.pp.ru) Received: (from ache@localhost) by nagual.pp.ru (8.9.1/8.9.1) id IAA22856; Sun, 6 Sep 1998 08:30:08 +0400 (MSD) (envelope-from ache) Message-ID: <19980906083007.A19996@nagual.pp.ru> Date: Sun, 6 Sep 1998 08:30:07 +0400 From: "Andrey A. Chernov" To: Mike Smith , current@FreeBSD.ORG Cc: dfr@FreeBSD.ORG Subject: Re: sbflush 2 panic, uipc_socket2.c Mail-Followup-To: Mike Smith , current@FreeBSD.org, dfr@freebsd.org References: <19980904103427.A3368@nagual.pp.ru> <199809040024.AAA01179@word.smith.net.au> <19980904143152.A20161@nagual.pp.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2i In-Reply-To: <19980904143152.A20161@nagual.pp.ru>; from ache@nagual.pp.ru on Fri, Sep 04, 1998 at 02:31:52PM +0400 Organization: Biomechanoid Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Fri, Sep 04, 1998 at 02:31:52PM +0400, Andrey A. Chernov wrote: > On Fri, Sep 04, 1998 at 12:24:25AM +0000, Mike Smith wrote: > > > On Thu, Sep 03, 1998 at 01:15:02PM +0000, Mike Smith wrote: > > > > > Approx. once a day I have this panic (see subj.) on heavily loaded > > > > > -current HTTP server. Is any solution possible? Briefly looking through > > > > > code I think some sort of spl protection needed between freeing sb_cc and > > > > > checking it again. > > > > > > > > That sounds reasonable. What have you tried so far? > > > > > > No, just gather more panic details: > > > > How about whacking an splnet() around it and letting us know how it > > goes? > > No, the code already under splnet() from tcp_usr_disconnect() > It means that internal mbuf structure is changed somehow under splnet() I found that the panic trigger is new ip defragmenting code from dfr, there is no such panic before this commit. ssh key exchange (large packets) can effectively cause panic after 10min when someone quickly log in with ssh after machine come after reboot. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ MTH/SH/HE S-- W-- N+ PEC>+ D A a++ C G>+ QH+(++) 666+>++ Y To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message