From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marc MERLIN Subject: Re: 4.8.8, bcache deadlock and hard lockup Date: Sun, 20 Nov 2016 16:04:13 -0800 Message-ID: <20161121000413.n5okcoqthaagiqel@merlins.org> References: <20161118164643.g7ttuzgsj74d6fbz@merlins.org> <20161118184915.j6dlazbgminxnxzx@merlins.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from magic.merlins.org ([209.81.13.136]:50771 "EHLO mail1.merlins.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752029AbcKUAES (ORCPT ); Sun, 20 Nov 2016 19:04:18 -0500 Content-Disposition: inline In-Reply-To: Sender: linux-bcache-owner@vger.kernel.org List-Id: linux-bcache@vger.kernel.org To: Coly Li Cc: linux-bcache@vger.kernel.org On Mon, Nov 21, 2016 at 05:13:41AM +0800, Coly Li wrote: > Hi Marc, > > Could you please give me an exact procedure that how the environment is > configured, and how the commands are executed ? I will try them on my > machine to reproduce it. Well, let's see: source and destination look like this: mdadm raid 5 \ ssd bcache ---- dmcrypt - btrfs Then I do a 10TB btrfs send from dev1 to dev2 The exact commands, some I typed a long time ago. This is live data, not a test environment I just spin up for testing. That said, is that what you were asking about? But as I said, right now I removed the caching device from dev2, so bcache is still there but doing nothing, and I'm still getting kernel OOM problems as per http://pastebin.com/2LUicF3k I'm currently testing 4.19rc5 to see if the VM changes help my case. Marc -- "A mouse is a device used to point at the xterm you want to type in" - A.S.R. Microsoft is to operating systems .... .... what McDonalds is to gourmet cooking Home page: http://marc.merlins.org/ | PGP 1024R/763BE901