mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Phy Prabab <phyprabab@yahoo.com>
To: Lars Marowsky-Bree <lmb@suse.de>, linux-kernel@vger.kernel.org
Subject: Re: md and multipathing
Date: Sat, 30 Oct 2004 12:08:50 -0700 (PDT)	[thread overview]
Message-ID: <20041030190850.23015.qmail@web51808.mail.yahoo.com> (raw)
In-Reply-To: <20041030174802.GK32712@marowsky-bree.de>

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 2137 bytes --]

Lars et all,

This what I am doing I believe.  Here is the command:

mdadm --create --force -lmp -n2 /dev/md0 /dev/sda
/dev/sdc

First controller is /dev/sda /dev/sdb (channel a and b
respective) and the second controller is /dev/sdc
/dev/sdd (again, channel a and b respecitve).

Here I am using multipath (enabled in kernel,
268,269,2610rc1) with the two paths to the FC volume.
I have found that it only drives one channel of the md
device as reported by vmstat.  More over, the traffic
appears to only go to the first device listed e.g. if
I change the ordering and use channel "b" first, than
all traffic to the md is driven on that channel.

Put another way, this is what I believe I have done:
device 0:0:0:0 + 1:0:0:0 = md0
device 0:0:0:1 + 1:0:0:1 = md1

scsi dev sda + sdc = md0
scsi dev sdb + sdd = md1

So have I done something wrong with the setup?  Am I
to question mdadm or md device?  

I get the exact same behavior as if I were to make the
device have a sprae path i.e. mdadm --create --force
-lmp -n1 -x1 /dev/md0 /dev/sda /dev/sdc,  but without
the failover device.  What I would want is to have
traffice be "exqually balanced" across both channels
(e.g. device paths /dev/sda /dev/sdc) with hopes that
this will improve my throughput.

Thanks!
Phy
--- Lars Marowsky-Bree <lmb@suse.de> wrote:

> On 2004-10-28T17:25:02, Phy Prabab
> <phyprabab@yahoo.com> wrote:
> 
> > I have a question concerning md driver: is there a
> way
> > to have a multipath md that is mulitplexed?
> 
> With 2.6, use the Device-Mapper multipath module.
> 
> 
> Sincerely,
>     Lars Marowsky-Brée <lmb@suse.de>
> 
> -- 
> High Availability & Clustering
> SUSE Labs, Research and Development
> SUSE LINUX AG - A Novell company
> 
> -
> To unsubscribe from this list: send the line
> "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at 
> http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 



		
__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail 

      reply	other threads:[~2004-10-30 19:09 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-10-29  0:25 Phy Prabab
2004-10-30 17:48 ` Lars Marowsky-Bree
2004-10-30 19:08   ` Phy Prabab [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=20041030190850.23015.qmail@web51808.mail.yahoo.com \
    --to=phyprabab@yahoo.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lmb@suse.de \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®