From: "Bryn M. Reeves" <bmr@redhat.com>
To: device-mapper development <dm-devel@redhat.com>
Cc: vgoyal@redhat.com
Subject: Re: Device mapper target driver for Write tracking or Write split
Date: Fri, 9 Jan 2015 15:13:38 +0000 [thread overview]
Message-ID: <20150109151337.GA17009@localhost.localdomain> (raw)
In-Reply-To: <a6164707-a0dc-426d-9324-a5fba2170682@default>
On Fri, Jan 09, 2015 at 06:51:28AM -0800, Suresh Babu Kandukuru wrote:
> 1) Basically we map the LUN from storage arrary to one of linux host . Linux host sees that LUN with mutlipath ( dm-multipath) abstact device /dev/mapper/LUNID . Linux host can mount that lun and do some read and write operation on that . We want to cature those writes and send those writes to another appliance . we will use that appilance for CDP( continues data protection) of that LUN .
The closest to this is probably dm-replicator:
http://people.redhat.com/heinzm/sw/dm/dm-replicator/src/linux/Documentation/device-mapper/replicator.txt
However it's not upstream and is not currently under development.
In addition to the single device write replication you discuss it supports
a range of device and link counts and types (e.g. the README above shows
an example with multiple local and remote devices and inter site links).
> 2) My idea is To track writes on the mapped lun in Linux, Can we instantiate a write tracking Device Mapper device stacked on top of the mapped lun device ( /dev/mapper/LUIN) ?. Then, whenever the write tracking Device Mapper device receives write request, it tracks the write. After that, it passes the write request down to the underlying lun device. In addition, to track writes to a Lun device, Linux host needs to be told to access the write tracking device instead of the LUN device itself.
In terms of targets that are upstream and supported currently this has
some similarity to the mirroring modes now offered via the dm-raid target.
This exposes the Linux MD RAID personalities as device-mapper targets.
Some combination of RAID1 and write-behind behaviour for the replicated
device may give you what you're looking for.
Regards,
Bryn.
next prev parent reply other threads:[~2015-01-09 15:13 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-09 5:23 Device mapper target driver for Write tracking or Write split Suresh Babu Kandukuru
2015-01-09 14:23 ` Vivek Goyal
2015-01-09 14:51 ` Suresh Babu Kandukuru
2015-01-09 15:13 ` Bryn M. Reeves [this message]
2015-01-09 15:51 ` Suresh Babu Kandukuru
2015-01-09 16:59 ` Bryn M. Reeves
2015-01-12 10:04 ` Suresh Babu Kandukuru
2015-01-12 9:34 ` Zdenek Kabelac
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=20150109151337.GA17009@localhost.localdomain \
--to=bmr@redhat.com \
--cc=dm-devel@redhat.com \
--cc=vgoyal@redhat.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.