From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 11 Jan 2001 13:22:27 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 11 Jan 2001 13:22:18 -0500 Received: from Hell.WH8.TU-Dresden.De ([141.30.225.3]:5392 "EHLO Hell.WH8.TU-Dresden.De") by vger.kernel.org with ESMTP id ; Thu, 11 Jan 2001 13:22:03 -0500 Message-ID: <3A5DF9CC.2F614F2A@Hell.WH8.TU-Dresden.De> Date: Thu, 11 Jan 2001 19:22:04 +0100 From: "Udo A. Steinberg" Organization: Dept. Of Computer Science, Dresden University Of Technology X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.0-ac6 i686) X-Accept-Language: en, de-DE MIME-Version: 1.0 To: Linux Kernel Subject: Strange umount problem in latest 2.4.0 kernels Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org As previously reported by someone, there are occasional problems when shutting down with unmounting partitions, that are reported as busy for strange reasons. Keith Owens said it was supposedly a Redhat shutdown script issue and I since I'm not using Redhat, it's most likely not that. Upon fscking after reboot, I always have errors on a single inode and it's always the same one: /dev/hdb1: Inode 522901, i_blocks is 64, should be 8. FIXED Can someone tell me an easy and reliable way of figuring out which file (program) uses said inode? I think that's probably the key to figuring out why the partition is busy on umount. -Udo. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/