From mboxrd@z Thu Jan 1 00:00:00 1970 From: Neil Brown Subject: Re: Implementing low level timeouts within MD Date: Tue, 30 Oct 2007 15:47:43 +1100 Message-ID: <18214.46959.424946.683169@notabene.brown> References: <1193418753.4771.17.camel@w100> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: message from Alberto Alonso on Friday October 26 Sender: linux-raid-owner@vger.kernel.org To: Alberto Alonso Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids On Friday October 26, alberto@ggsys.net wrote: > I've been asking on my other posts but haven't seen > a direct reply to this question: > > Can MD implement timeouts so that it detects problems when > drivers don't come back? No. However it is possible that we will start sending the BIO_RW_FAILFAST flag down on some or all requests. That might make drivers fail more promptly, which might be good thing. However it won't fix bugs in drivers and - as has been said elsewhere on this thread - that is the real problem. NeilBrown