Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 06 May 2003 15:06:33 -0400
From:      Bill Moran <wmoran@potentialtech.com>
To:        YOU <trodat@server1.ultratrends.com>
Cc:        questions@freebsd.org
Subject:   Re: Realtime Filesystem Replication
Message-ID:  <3EB807B9.4070603@potentialtech.com>
In-Reply-To: <Pine.BSF.4.21.0305061221200.47177-100000@server1.ultratrends.com>
References:  <Pine.BSF.4.21.0305061221200.47177-100000@server1.ultratrends.com>

next in thread | previous in thread | raw e-mail | index | archive | help
YOU wrote:
> Thanks so far to the suggestions including rsync and unison. Both appear
> to be triggered upon a command line or user typed command. Is someone
> using a system that tracks the mtimes for files and updates without
> prompt? 

Are you sure you really need _realtime_?  That's a pretty tall order, and
I don't know of anything that can provide it.

OTOH, with rsync as a cron job, you can easily mirror a large amount of
data, say, every 5 minutes.  If a 5-minute lag in data replication is
acceptable, I would recommend using rsync instead of looking for something
more "real time".  I don't know if any true "real time" synchronizing
exists.

Let me ask you a few questions: 1. What is the maximum acceptable "lag" for
data replication?  2. How much data do you estimate there will be?  (both
megs and # of files) 3. How often do you estimate the data will change?

If 1 is longer than a few minutes, and 2 and 3 aren't terribly demanding,
rsync is probably the way to go.

On a related note: I had an idea that you should be able to create some
sort of network RAID using the new GEOM system.  Something that would allow
everything that was written to a local filsystem to also be mirrored via
NFS to another system.  I haven't had a chance to do anything more than
dream about it, though.

-- 
Bill Moran
Potential Technologies
http://www.potentialtech.com



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3EB807B9.4070603>