From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 14 Feb 2003 19:31:13 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 14 Feb 2003 19:31:13 -0500 Received: from packet.digeo.com ([12.110.80.53]:61317 "EHLO packet.digeo.com") by vger.kernel.org with ESMTP id ; Fri, 14 Feb 2003 19:31:12 -0500 Date: Fri, 14 Feb 2003 16:39:18 -0800 From: Andrew Morton To: Siim Vahtre Cc: linux-kernel@vger.kernel.org Subject: Re: swap never cleaned up Message-Id: <20030214163918.15a7209f.akpm@digeo.com> In-Reply-To: References: X-Mailer: Sylpheed version 0.8.9 (GTK+ 1.2.10; i586-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 15 Feb 2003 00:40:48.0520 (UTC) FILETIME=[E2308080:01C2D48A] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Siim Vahtre wrote: > > Hi. > > Since moving from 2.5.59 to 2.5.60 I've noticed some strange > behaviour with swap managment. For some unknown reason the > swapspace starts to fill up but it will NEVER get freed. > Hence, it is first time in my life when I actually see swap > used more than 20M on this computer! Everything seems OK here, from a quick swap and tmpfs test. Is it possible that some application is leaking memory, and that swapoff causes it to be oom-killed? Take a careful look at the process listing, see if some process is using a lot of memory. Also the contents of /proc/meminfo. And see if you can work out what operation is causing this. Thanks.