From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262322AbVG2EZt (ORCPT ); Fri, 29 Jul 2005 00:25:49 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262324AbVG2EZt (ORCPT ); Fri, 29 Jul 2005 00:25:49 -0400 Received: from mail.dvmed.net ([216.237.124.58]:20430 "EHLO mail.dvmed.net") by vger.kernel.org with ESMTP id S262322AbVG2EZr (ORCPT ); Fri, 29 Jul 2005 00:25:47 -0400 Message-ID: <42E9AFC6.9010805@pobox.com> Date: Fri, 29 Jul 2005 00:25:42 -0400 From: Jeff Garzik User-Agent: Mozilla Thunderbird 1.0.6-1.1.fc4 (X11/20050720) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Yaroslav Halchenko CC: linux kernel mailing list Subject: Re: 2.6.8 -> 2.6.11 (+ata-dev patch) -- HDD is always on References: <20050729041031.GU16285@washoe.onerussian.com> In-Reply-To: <20050729041031.GU16285@washoe.onerussian.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Score: 0.0 (/) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Yaroslav Halchenko wrote: > I've been running debian stable with 2.6.8 kernel but due to recent > failure of the SATA harddrive I've decided to upgrade to 2.6.11 + libata > patch (2.6.11-libata-dev1.patch.gz) and recent smartmontools > > After everything worked out and I decided to rest in piece but I've > found that HDD LED light nomater what. It seems to turn off during BIOS > checks and then kicks in 1-2 secs after kernel starts booting. No > prominent harddrive activity noise can be heard but this drive is > quite silent so it is hard to say. Does this happen in unpatched 2.6.12.3 or 2.6.13-rc4? > QUESTION: > > LED constantly ON - does it signal about a problem or may be just > that some status bit is hanging? Should I worry and try differen kernel > version? > > YIKES: ran hddtemp /dev/sda and whole box hanged... SysRq keys - no > effect... heh heh... reboot -> nothing in logs > P.S. was ata-dev patch incorporated in recent kernel versions so I could > use SMART with vanilla kernel? This is one of the reasons why SMART support is not yet in the mainline kernel! Jeff