From: Guoqing Jiang <gqjiang@suse.com>
To: shli@kernel.org, linux-raid@vger.kernel.org
Cc: Shaohua Li <shli@fb.com>, NeilBrown <neilb@suse.com>
Subject: Re: [PATCH 3/3] MD: hold mddev lock for md-cluster receive thread
Date: Mon, 1 Aug 2016 16:38:59 +0800 [thread overview]
Message-ID: <579F0AA3.5090806@suse.com> (raw)
In-Reply-To: <7763e508fb97d44bd61e826912055617b8be2c2d.1469922791.git.shli@fb.com>
Hi,
On 07/31/2016 07:54 AM, shli@kernel.org wrote:
> From: Shaohua Li <shli@fb.com>
>
> md-cluster receive thread calls .quiesce too, let it hold mddev lock.
I'd suggest hold on for the patchset, I can find lock problem easily with
the patchset applied. Take a resyncing clusteed raid1 as example.
md127_raid1 thread held reconfig_mutex then update sb, so it needs dlm
token lock. Meanwhile md127_resync thread got token lock and wants
EX on ack lock but recv_daemon can't release ack lock since recv_daemon
doesn't get reconfig_mutex.
etalinux135:~ # ps aux|grep md|grep D
root 2028 0.0 0.0 0 0 ? D 16:24 0:00
[md127_raid1]
root 2041 0.0 0.0 0 0 ? D 16:24 0:00
[md127_resync]
betalinux135:~ # cat /proc/2028/stack
[<ffffffffa05f2660>] metadata_update_start+0xa0/0xb0 [md_cluster]
[<ffffffffa06cb1ce>] md_update_sb.part.50+0x8e/0x810 [md_mod]
[<ffffffffa06ccb8c>] md_check_recovery+0x23c/0x4f0 [md_mod]
[<ffffffffa06f6312>] raid1d+0x42/0x7d0 [raid1]
[<ffffffffa06c6d10>] md_thread+0x130/0x150 [md_mod]
[<ffffffff810995ed>] kthread+0xbd/0xe0
[<ffffffff815e96bf>] ret_from_fork+0x3f/0x70
[<ffffffff81099530>] kthread+0x0/0xe0
[<ffffffffffffffff>] 0xffffffffffffffff
betalinux135:~ # cat /proc/2041/stack
[<ffffffffa05f24cb>] dlm_lock_sync+0x6b/0x80 [md_cluster]
[<ffffffffa05f2708>] __sendmsg+0x98/0x130 [md_cluster]
[<ffffffffa05f282d>] sendmsg+0x1d/0x30 [md_cluster]
[<ffffffffa05f2b31>] resync_info_update+0x81/0xb0 [md_cluster]
[<ffffffffa06f3ad7>] sync_request+0xa57/0xaf0 [raid1]
[<ffffffffa06ca2dd>] md_do_sync+0x90d/0xe80 [md_mod]
[<ffffffffa06c6d10>] md_thread+0x130/0x150 [md_mod]
[<ffffffff810995ed>] kthread+0xbd/0xe0
[<ffffffff815e96bf>] ret_from_fork+0x3f/0x70
[<ffffffff81099530>] kthread+0x0/0xe0
[<ffffffffffffffff>] 0xffffffffffffffff
Thanks,
Guoqing
next prev parent reply other threads:[~2016-08-01 8:38 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-30 23:54 [PATCH 1/3] MD: hold mddev lock for .quiesce in md_do_sync shli
2016-07-30 23:54 ` [PATCH 2/3] MD: hold mddev lock to change bitmap location shli
2016-08-03 0:03 ` NeilBrown
2016-07-30 23:54 ` [PATCH 3/3] MD: hold mddev lock for md-cluster receive thread shli
2016-08-01 8:38 ` Guoqing Jiang [this message]
2016-08-01 21:45 ` Shaohua Li
2016-08-02 9:52 ` Guoqing Jiang
2016-08-02 22:44 ` Shaohua Li
2016-08-03 3:18 ` Guoqing Jiang
2016-08-03 0:09 ` NeilBrown
2016-08-03 3:42 ` Guoqing Jiang
2016-07-31 6:03 ` [PATCH 1/3] MD: hold mddev lock for .quiesce in md_do_sync yizhan
2016-08-02 23:47 ` NeilBrown
2016-08-04 3:16 ` NeilBrown
2016-08-06 4:14 ` Shaohua Li
2016-08-12 0:04 ` NeilBrown
2016-08-17 1:28 ` Shaohua Li
2016-08-24 4:49 ` NeilBrown
2016-08-24 5:25 ` Shaohua Li
2016-08-25 4:59 ` NeilBrown
2016-08-25 17:17 ` Shaohua Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=579F0AA3.5090806@suse.com \
--to=gqjiang@suse.com \
--cc=linux-raid@vger.kernel.org \
--cc=neilb@suse.com \
--cc=shli@fb.com \
--cc=shli@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.