All of lore.kernel.org
 help / color / mirror / Atom feed
From: Martin Wilck <mwilck@suse.com>
To: Guan Junxiong <guanjunxiong@huawei.com>
Cc: dm-devel@redhat.com, chengjike.cheng@huawei.com,
	xose.vazquez@gmail.com, niuhaoxin@huawei.com,
	shenhong09@huawei.com
Subject: Re: [PATCH] multipath-tools: support for coalescing heterogenous paths
Date: Mon, 04 Sep 2017 23:09:48 +0200	[thread overview]
Message-ID: <1504559388.4605.17.camel@suse.com> (raw)
In-Reply-To: <de4c2d55-4556-d576-75f9-b5fef5212a32@huawei.com>

On Sat, 2017-09-02 at 14:43 +0800, Guan Junxiong wrote:
> On 2017/8/22 22:21, Guan Junxiong wrote:
> > Hi, guys,
> > 
> > Any comments about this patch?
> > 
> > IMO, this feature of coalescing heterogenous paths without
> > interrupting
> > uplayer transaction is worthy to integrate into multipath-tools.
> > Please
> > consider about it.
> > 
> > I am glad to see if there is anything which can be improve for this
> > patch.

I admit I haven't looked into the details this time. But I spoke with
Hannes about it, and we agreed that this is not something we would like
to see in upstream multipath-tools. multipath is taking efforts to make
sure that we do *not* coalesce any devices that aren't actually paths
to the same disk. Your patch is trying to do the opposite.

AFAICS this is a special feature provided by certain storage arrays
from Huawei. It doesn't work at all without the storage array being
properly configured to take appropriate action to keep the
"heterogeneous paths" in sync. We think support for this belongs into
Huawei software for the time being. If this feature becomes
standardized in the future, and if the OS has some reliable,
standardized means to detect this feature in the array, it might be a
different issue. But even then, we'd need to use extreme caution not to
corrupt our users' data.

This is Hannes' and my opinion, others feel free to speak up if you
disagree.

Martin

> > 
> > Thanks
> > Guan Junxiong
> > 
> > 
> > On 2017/8/17 16:15, Guan Junxiong wrote:
> > > This patch supports coalescing heterogenous paths by using
> > > udev rules files. This is useful in the scenario of migrating
> > > data for heterogenous arrays without interrupting uplayer
> > > transaction. The udev rules file, which is located in the
> > > /etc/udev/rules.d/ directory, is automatically generated by
> > > multipath with new options -m and -n.
> > > 
> > > For example:
> > > multipath -m dm-0 -n  1000036006016028
> > > 
> > > The geneated udev rules is as follows:
> > > KERNEL=="sd*[!0-9]|nvme*[0-9]n*[0-9]", ENV{DEVTYPE}=="disk",
> > > ENV{ID_SERIAL}=="1000036006016028", ENV{ID_SERIAL}="36006016-
> > > 028f03600588db521dc7ce71"
> > > 
> > > KERNEL=="sd*[!0-9]|nvme*[0-9]n*[0-9]", ENV{DEVTYPE}=="disk",
> > > ENV{ID_WWN}=="1000036006016028", ENV{ID_WWN}="36006016028f036-
> > > 00588db521dc7ce71"
> > > 
> > > KERNEL=="sd*[!0-9]|nvme*[0-9]n*[0-9]", ENV{DEVTYPE}=="disk",
> > > ENV{ID_UUID}=="1000036006016028", ENV{ID_UUID}="36006016028f03-
> > > 600588db521dc7ce71"
> > > 
> > > Signed-off-by: Junxiong Guan <guanjunxiong@huawei.com>
> > > ---
> > >  libmultipath/Makefile   |   3 +-
> > >  libmultipath/config.h   |   1 +
> > >  libmultipath/udevrule.c | 191
> > > ++++++++++++++++++++++++++++++++++++++++++++++++
> > >  libmultipath/udevrule.h |  13 ++++
> > >  multipath/main.c        |  29 +++++++-
> > >  multipath/multipath.8   |  12 +++
> > >  6 files changed, 247 insertions(+), 2 deletions(-)
> > >  create mode 100644 libmultipath/udevrule.c
> > >  create mode 100644 libmultipath/udevrule.h
> 
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

-- 
Dr. Martin Wilck <mwilck@suse.com>, Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel

      reply	other threads:[~2017-09-04 21:09 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-17  8:15 [PATCH] multipath-tools: support for coalescing heterogenous paths Guan Junxiong
2017-08-22 14:21 ` Guan Junxiong
2017-09-02  6:43   ` Guan Junxiong
2017-09-04 21:09     ` Martin Wilck [this message]

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=1504559388.4605.17.camel@suse.com \
    --to=mwilck@suse.com \
    --cc=chengjike.cheng@huawei.com \
    --cc=dm-devel@redhat.com \
    --cc=guanjunxiong@huawei.com \
    --cc=niuhaoxin@huawei.com \
    --cc=shenhong09@huawei.com \
    --cc=xose.vazquez@gmail.com \
    /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.