From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.5 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id F19A0ECDE27 for ; Wed, 11 Sep 2019 06:55:48 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id BA01221928 for ; Wed, 11 Sep 2019 06:55:48 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726724AbfIKGzs (ORCPT ); Wed, 11 Sep 2019 02:55:48 -0400 Received: from mx2.suse.de ([195.135.220.15]:46318 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726657AbfIKGzr (ORCPT ); Wed, 11 Sep 2019 02:55:47 -0400 X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (unknown [195.135.220.254]) by mx1.suse.de (Postfix) with ESMTP id 07131ABC7; Wed, 11 Sep 2019 06:55:44 +0000 (UTC) Subject: Re: Mount/df/PAM login hangs during rsync to btrfs subvolume, or maybe doing btrfs subvolume snapshot To: David Newall , linux-btrfs@vger.kernel.org References: From: Nikolay Borisov Openpgp: preference=signencrypt Autocrypt: addr=nborisov@suse.com; prefer-encrypt=mutual; keydata= mQINBFiKBz4BEADNHZmqwhuN6EAzXj9SpPpH/nSSP8YgfwoOqwrP+JR4pIqRK0AWWeWCSwmZ T7g+RbfPFlmQp+EwFWOtABXlKC54zgSf+uulGwx5JAUFVUIRBmnHOYi/lUiE0yhpnb1KCA7f u/W+DkwGerXqhhe9TvQoGwgCKNfzFPZoM+gZrm+kWv03QLUCr210n4cwaCPJ0Nr9Z3c582xc bCUVbsjt7BN0CFa2BByulrx5xD9sDAYIqfLCcZetAqsTRGxM7LD0kh5WlKzOeAXj5r8DOrU2 GdZS33uKZI/kZJZVytSmZpswDsKhnGzRN1BANGP8sC+WD4eRXajOmNh2HL4P+meO1TlM3GLl EQd2shHFY0qjEo7wxKZI1RyZZ5AgJnSmehrPCyuIyVY210CbMaIKHUIsTqRgY5GaNME24w7h TyyVCy2qAM8fLJ4Vw5bycM/u5xfWm7gyTb9V1TkZ3o1MTrEsrcqFiRrBY94Rs0oQkZvunqia c+NprYSaOG1Cta14o94eMH271Kka/reEwSZkC7T+o9hZ4zi2CcLcY0DXj0qdId7vUKSJjEep c++s8ncFekh1MPhkOgNj8pk17OAESanmDwksmzh1j12lgA5lTFPrJeRNu6/isC2zyZhTwMWs k3LkcTa8ZXxh0RfWAqgx/ogKPk4ZxOXQEZetkEyTFghbRH2BIwARAQABtCNOaWtvbGF5IEJv cmlzb3YgPG5ib3Jpc292QHN1c2UuY29tPokCOAQTAQIAIgUCWIo48QIbAwYLCQgHAwIGFQgC CQoLBBYCAwECHgECF4AACgkQcb6CRuU/KFc0eg/9GLD3wTQz9iZHMFbjiqTCitD7B6dTLV1C ddZVlC8Hm/TophPts1bWZORAmYIihHHI1EIF19+bfIr46pvfTu0yFrJDLOADMDH+Ufzsfy2v HSqqWV/nOSWGXzh8bgg/ncLwrIdEwBQBN9SDS6aqsglagvwFD91UCg/TshLlRxD5BOnuzfzI Leyx2c6YmH7Oa1R4MX9Jo79SaKwdHt2yRN3SochVtxCyafDlZsE/efp21pMiaK1HoCOZTBp5 VzrIP85GATh18pN7YR9CuPxxN0V6IzT7IlhS4Jgj0NXh6vi1DlmKspr+FOevu4RVXqqcNTSS E2rycB2v6cttH21UUdu/0FtMBKh+rv8+yD49FxMYnTi1jwVzr208vDdRU2v7Ij/TxYt/v4O8 V+jNRKy5Fevca/1xroQBICXsNoFLr10X5IjmhAhqIH8Atpz/89ItS3+HWuE4BHB6RRLM0gy8 T7rN6ja+KegOGikp/VTwBlszhvfLhyoyjXI44Tf3oLSFM+8+qG3B7MNBHOt60CQlMkq0fGXd mm4xENl/SSeHsiomdveeq7cNGpHi6i6ntZK33XJLwvyf00PD7tip/GUj0Dic/ZUsoPSTF/mG EpuQiUZs8X2xjK/AS/l3wa4Kz2tlcOKSKpIpna7V1+CMNkNzaCOlbv7QwprAerKYywPCoOSC 7P25Ag0EWIoHPgEQAMiUqvRBZNvPvki34O/dcTodvLSyOmK/MMBDrzN8Cnk302XfnGlW/YAQ csMWISKKSpStc6tmD+2Y0z9WjyRqFr3EGfH1RXSv9Z1vmfPzU42jsdZn667UxrRcVQXUgoKg QYx055Q2FdUeaZSaivoIBD9WtJq/66UPXRRr4H/+Y5FaUZx+gWNGmBT6a0S/GQnHb9g3nonD jmDKGw+YO4P6aEMxyy3k9PstaoiyBXnzQASzdOi39BgWQuZfIQjN0aW+Dm8kOAfT5i/yk59h VV6v3NLHBjHVw9kHli3jwvsizIX9X2W8tb1SefaVxqvqO1132AO8V9CbE1DcVT8fzICvGi42 FoV/k0QOGwq+LmLf0t04Q0csEl+h69ZcqeBSQcIMm/Ir+NorfCr6HjrB6lW7giBkQl6hhomn l1mtDP6MTdbyYzEiBFcwQD4terc7S/8ELRRybWQHQp7sxQM/Lnuhs77MgY/e6c5AVWnMKd/z MKm4ru7A8+8gdHeydrRQSWDaVbfy3Hup0Ia76J9FaolnjB8YLUOJPdhI2vbvNCQ2ipxw3Y3c KhVIpGYqwdvFIiz0Fej7wnJICIrpJs/+XLQHyqcmERn3s/iWwBpeogrx2Lf8AGezqnv9woq7 OSoWlwXDJiUdaqPEB/HmGfqoRRN20jx+OOvuaBMPAPb+aKJyle8zABEBAAGJAh8EGAECAAkF AliKBz4CGwwACgkQcb6CRuU/KFdacg/+M3V3Ti9JYZEiIyVhqs+yHb6NMI1R0kkAmzsGQ1jU zSQUz9AVMR6T7v2fIETTT/f5Oout0+Hi9cY8uLpk8CWno9V9eR/B7Ifs2pAA8lh2nW43FFwp IDiSuDbH6oTLmiGCB206IvSuaQCp1fed8U6yuqGFcnf0ZpJm/sILG2ECdFK9RYnMIaeqlNQm iZicBY2lmlYFBEaMXHoy+K7nbOuizPWdUKoKHq+tmZ3iA+qL5s6Qlm4trH28/fPpFuOmgP8P K+7LpYLNSl1oQUr+WlqilPAuLcCo5Vdl7M7VFLMq4xxY/dY99aZx0ZJQYFx0w/6UkbDdFLzN upT7NIN68lZRucImffiWyN7CjH23X3Tni8bS9ubo7OON68NbPz1YIaYaHmnVQCjDyDXkQoKC R82Vf9mf5slj0Vlpf+/Wpsv/TH8X32ajva37oEQTkWNMsDxyw3aPSps6MaMafcN7k60y2Wk/ TCiLsRHFfMHFY6/lq/c0ZdOsGjgpIK0G0z6et9YU6MaPuKwNY4kBdjPNBwHreucrQVUdqRRm RcxmGC6ohvpqVGfhT48ZPZKZEWM+tZky0mO7bhZYxMXyVjBn4EoNTsXy1et9Y1dU3HVJ8fod 5UqrNrzIQFbdeM0/JqSLrtlTcXKJ7cYFa9ZM2AP7UIN9n1UWxq+OPY9YMOewVfYtL8M= Message-ID: <1a651f17-ba40-2f17-403e-69999e927b2d@suse.com> Date: Wed, 11 Sep 2019 09:55:43 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.8.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org On 11.09.19 г. 9:45 ч., David Newall wrote: > Hi All, > > I might have misunderstood how to report a problem.  I registered for > bugzilla and reported a bug > (https://bugzilla.kernel.org/show_bug.cgi?id=204757), but, perhaps I > should have sent this message to this mailing list, first.  My apologies > if I bungled it. > > I've been trying to track down a problem, intermittently, for a long > time, and now need to reach out for advice.  I apologise in advance for > the quality of this report, which I feel includes more detail than > needed, yet may be missing what's important.  I'm trying my best. > > The brief summary is that my system hangs during SSH login while a > backup is in progress.  Sshd uses PAM authentication.  The problem seems > to be related to mounts as df and mount also hang. > > The longer details are:  I'm running Ubuntu 16.04.5 on a 64-bit VM under > kvm.  I backup data using the following steps: > > 1. Take an LVM2 snapshot of the (non-root) ext2 file-system mounted as > /data; > 2. Mount a btrfs file system as /backup; > 2. Mount the snapshot over an empty directory (may be subvolume; does it > make a difference?) on /backup/snapshot; > 3. Rsync the snapshot (with --archive --one-file-system --hard-links > --inplace --numeric-ids --delete) to a subvolume /backup/data (thus it > always contains /data as at last backup); > 4. Take btrfs subvolume snapshot of /backup/data; > 5. Unmount /backup/snapshot and /backup. > > By the time I get called, SSH logins via PAM hang (but complete > "immediately" if I re-configure sshd for UsePAM no).  Sessions which are > still logged in seem unaffected, except df and mount both hang.  I don't > know what else hangs. > > During all of these steps, the /data is almost static, maybe even be > completely static. > > I've queried my user, carefully, to determine the exact step where it > starts to hang, and am 90% confident in her answer, which indicates that > the hang-condition starts during rsync. > > Processes that were hanging complete normally when subvolume snapshot > finishes. > > There's a chance that processes complete when the snapshot or btrfs > file-system is unmounted, but I think it's before then because I've > tried running each step by hand, was unable to reproduce the problem, > probably because the amount of data to rsync in real-use is much larger > than I tried writing during that test.  At any rate, during that test I > could log in between and during each step of the procedure. > > The only messages in dmesg are "mounting ext2 file system using the ext4 > subsystem" and "mounted filesystem without journal. Opts: (null)", which > sounds right as I use "mount" instead of "mount -text2". > > When I tried running df under strace, strace's output was: > >   open("/proc/self/mountinfo", O_RDONLY)  = 3 >   fstat(3, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0 >   read(3, "18 24 0:17 / /sys rw,nosuid,node"..., 1024) = 1024 >   read(3, "ystemd/systemd-cgroups-agent,nam"..., 1024) = 1024 >   read(3, "t rw,nosuid,nodev,noexec,relatim"..., 1024) = 1024 >   read(3, > > After the subvolume snapshot completed, strace continued producing output: > >   "fs lxcfs rw,user_id=0,group_id=0"..., 1024) = 624 >   --- SIGCONT {si_signo=SIGCONT, si_code=SI_USER, si_pid=28055, > si_uid=1000} --- >   read(3, "", 1024)                       = 0 >   lseek(3, 0, SEEK_CUR)                   = 3696 >   close(3)                                = 0 > > I think the SIGCONT was because I suspended the parent, strace, using > Ctrl-Z. > > I could just leave sshd doing non-PAM authentication but I think that's > the wrong approach.  How do I zero in on this problem? When the issue manifests do : echo w > /proc/sysrq-trigger This should provide a backtrace for all threads which are currently in uninterruptible sleep. If it's a deadlock due to btrfs being stuck we should see it. Also provide your exact kernel version. > > Thanks, > > David > >