From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wr0-f195.google.com ([209.85.128.195]:34774 "EHLO mail-wr0-f195.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751636AbdITQHt (ORCPT ); Wed, 20 Sep 2017 12:07:49 -0400 Date: Wed, 20 Sep 2017 18:07:35 +0200 From: Kent Overstreet To: Coly Li Cc: linux-bcache@vger.kernel.org, linux-block@vger.kernel.org, Nix , Kai Krakow , Eric Wheeler , Junhui Tang , stable@vger.kernel.org Subject: Re: [PATCHv2] bcache: option for allow stale data on read failure Message-ID: <20170920160735.jp4riq7x3qc472px@kmo-pixel> References: <20170919222433.24336-1-colyli@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20170919222433.24336-1-colyli@suse.de> Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org On Wed, Sep 20, 2017 at 06:24:33AM +0800, Coly Li wrote: > When bcache does read I/Os, for example in writeback or writethrough mode, > if a read request on cache device is failed, bcache will try to recovery > the request by reading from cached device. If the data on cached device is > not synced with cache device, then requester will get a stale data. > > For critical storage system like database, providing stale data from > recovery may result an application level data corruption, which is > unacceptible. But for some other situation like multi-media stream cache, > continuous service may be more important and it is acceptible to fetch > a chunk of stale data. > > This patch tries to solve the above conflict by adding a sysfs option > /sys/block/bcache/bcache/allow_stale_data_on_failure > which is defaultly cleared (to 0) as disabled. Now people can make choices > for different situations. IMO this is just a bug, I'd rather not have an option to keep the buggy behaviour. How about this patch: commit 2746f9c1f962288d8c5d7dabe698bf7b3fddd405 Author: Kent Overstreet Date: Wed Sep 20 18:06:37 2017 +0200 bcache: Don't recover from IO errors when reading dirty data Signed-off-by: Kent Overstreet diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index 382397772a..c2d57ef953 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -532,8 +532,10 @@ static int cache_lookup_fn(struct btree_op *op, struct btree *b, struct bkey *k) PTR_BUCKET(b->c, k, ptr)->prio = INITIAL_PRIO; - if (KEY_DIRTY(k)) + if (KEY_DIRTY(k)) { s->read_dirty_data = true; + s->recoverable = false; + } n = bio_next_split(bio, min_t(uint64_t, INT_MAX, KEY_OFFSET(k) - bio->bi_iter.bi_sector),