From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nix Subject: Re: 4.11.2: reshape raid5 -> raid6 atop bcache deadlocks at start on md_attr_store / raid5_make_request Date: Wed, 24 May 2017 20:42:48 +0100 Message-ID: <87d1ayhwhz.fsf@esperi.org.uk> References: <87lgppz221.fsf@esperi.org.uk> <87a865jf9a.fsf@notabene.neil.brown.name> <87fufwy3lr.fsf@esperi.org.uk> <87poezhwsa.fsf@notabene.neil.brown.name> Mime-Version: 1.0 Content-Type: text/plain Return-path: In-Reply-To: <87poezhwsa.fsf@notabene.neil.brown.name> (NeilBrown's message of "Wed, 24 May 2017 11:24:21 +1000") Sender: linux-raid-owner@vger.kernel.org To: NeilBrown Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids On 24 May 2017, NeilBrown said: > Alternately, this might do it. Bingo! Ticking away nicely: [ 147.538274] md/raid:md125: device sda3 operational as raid disk 0 [ 147.574356] md/raid:md125: device sdd3 operational as raid disk 3 [ 147.586482] md/raid:md125: device sdc3 operational as raid disk 2 [ 147.598571] md/raid:md125: device sdb3 operational as raid disk 1 [ 147.613949] md/raid:md125: raid level 6 active with 4 out of 5 devices, algorithm 18 [ 147.776155] md: reshape of RAID array md125 md125 : active raid6 sda3[0] sdf3[5] sdd3[4] sdc3[2] sdb3[1] 15391689216 blocks super 1.2 level 6, 512k chunk, algorithm 18 [5/4] [UUUU_] [>....................] reshape = 0.0% (2669056/5130563072) finish=1363.5min speed=62678K/sec (wow, this is a lot faster: only 1363 min / 70MiB/s, versus the --backup-file one, for an array half the size, that was >2500 min, 12MiB/s. I mean, yes, it was at the slow end of the disk, but they don't differ in speed *that* much.) Machine still perfectly responsive, as if on reshape was happening at all. (But then, I've got used to that. The last time I saw any md resync- induced delays was long ago when I had a sym53c75 in the loop with an I/O rate of 10MiB/s. I think delays are unavoidable with something that slow in there.)