From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from soda.linbit (office.linbit [86.59.100.100]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by mail.linbit.com (LINBIT Mail Daemon) with ESMTP id 8E0262DF6256 for ; Tue, 26 Feb 2008 13:49:20 +0100 (CET) Date: Tue, 26 Feb 2008 13:49:20 +0100 From: Lars Ellenberg To: drbd-dev@lists.linbit.com Subject: Re: [Drbd-dev] I/O can hang on primary synctarget after an io error. Message-ID: <20080226124920.GA5727@barkeeper1.linbit> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: List-Id: Coordination of development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Mon, Feb 25, 2008 at 04:53:35PM -0500, Montrose, Ernest wrote: > Lars, > We appear to have a009fc907a14f69026b32fbb48a4db6f1cdd5ecd. Reading > your response what I get is that we are guaranteed that if we return > early in drbd_end_write_sec() then someone else would have done the > dec_local near the end or an inc_local was never done? this "return early" does not happen. because almost no driver actually implemented partial completion notification in the first place, and if some driver should indeed implement it, it is guaranteed to eventually call us with the the bio comlpetely done. finally, you trigger the "io-error" by fault insertion, which directly calls the endio handler with bi_size == 0. > Hmmm... our testing was not done with the latest git stuff. > I will do some things with the latest. keep us posted. -- : Lars Ellenberg Tel +43-1-8178292-55 : : LINBIT Information Technologies GmbH Fax +43-1-8178292-82 : : Vivenotgasse 48, A-1120 Vienna/Europe http://www.linbit.com :