From owner-freebsd-current@freebsd.org Fri Apr 1 20:37:25 2016 Return-Path: Delivered-To: freebsd-current@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 2734DAECEAD for ; Fri, 1 Apr 2016 20:37:25 +0000 (UTC) (envelope-from truckman@FreeBSD.org) Received: from gw.catspoiler.org (unknown [IPv6:2602:304:b010:ef20::f2]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "gw.catspoiler.org", Issuer "gw.catspoiler.org" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id 098C51C19; Fri, 1 Apr 2016 20:37:25 +0000 (UTC) (envelope-from truckman@FreeBSD.org) Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2]) by gw.catspoiler.org (8.15.2/8.15.2) with ESMTP id u31KbEOP002826; Fri, 1 Apr 2016 13:37:18 -0700 (PDT) (envelope-from truckman@FreeBSD.org) Message-Id: <201604012037.u31KbEOP002826@gw.catspoiler.org> Date: Fri, 1 Apr 2016 13:37:14 -0700 (PDT) From: Don Lewis Subject: Re: CURRENT slow and shaky network stability To: ohartman@zedat.fu-berlin.de cc: imb@protected-networks.net, kmacy@freebsd.org, freebsd-current@freebsd.org In-Reply-To: <20160401192444.69b1da2a.ohartman@zedat.fu-berlin.de> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Apr 2016 20:37:25 -0000 On 1 Apr, O. Hartmann wrote: > Am Wed, 30 Mar 2016 00:28:39 -0700 (PDT) > Don Lewis schrieb: >> Do you use tmpfs? Anything stored in there will get stashed in inactive >> memory and/or swap. > [...] > > Yes, /var/run and /tmp are on tmpfs There is probably not enough stashed in /var/run to make a difference, but if there is a lot in /tmp, it could bloat the amount of inactive memory and cause other things to be pushed to swap. What does df /tmp say? Depending on the workload, you might find that the change committed in r280963 helps interactivity. I still can't explain the ssh connection timeout problems that you are seeing. What does ps lax report as the MWCHAN for the stuck ssh and/or sshd processes?