James Washer wrote:
The big issues with backups IMHO is getting a consistent snapshot. If you
start copying a large data file, and it changes after you've copied the
beginning, you're screwed.
Over the years, I've generally adopted a smash-and-silver backup plan. Namely, I run
mirrored during normal operation. To create a backup, I silver a 3rd mirror. Once the 3rd
mirror is silvered, The file system can be frozen momentarily, and the 3rd mirror is
seperated from the set. A backup is now made from this "smashed" silver. Repeat
the process for as many filesystems as necessary. To work 100%, it requires cooperation
from the filesystem/kernel (to some degree)
Other backup schemes I've used involved BCL's (Block changed logging), that
allow the admin to only backup up those blocks that have been updated since the
last master. This is a GREAT space saver for large data farms with low update
rates. I tend to work on larger multi-terrabyte systems, so minimizing backup
times is very important.
Are you planning on your backup scheme being able to handle "hot" backups?
This makes particular sense in open source software. With proprietary
backup solutions, I always leaned against block level incrementals such
as provided by veritas, because without their software, there was pretty
much no way to recover your data... This meant that things like license
management or lame bugs could end up biting you in the butt at the very
worst possible time. There have always been file/tar-based products out
there, but like you said, not so efficient.
- jim
On Sat, 23 Jul 2005 16:25:43 -0700 (PDT)
Sebastian Smith <[EMAIL PROTECTED]> wrote:
Hi All,
Over my short sys admin career I've found few backup packages I like...
and none I'd pay for. As such, I'm bent on creating my own system that is
flexible, scalable, portable (maybe), and robust.
I've been using a modified version of the "backup to a central backup
server with 7 day incremental" backup script from rsync's "examples" page
for over a year now:
http://samba.anu.edu.au/rsync/examples.html
It works great, and is easily augmented with other scripts to backup to
CD, DVD, tape, or whatever. It doesn't offer a restore function, but,
since it stores backups as exact copies of existing files (the ideal
method... IMHO), replacing files is easy.
The primary problem I've found with this script is that it has to be
modified for each installation (effects a lot of the qualities I'm looking
for in backup software). So... I've started generalizing the script, and
have begun adding features I want... which are:
- Multi-source, and Multi-destination (including to/from remote hosts via
ssl tunnel)
- Optional Parallelized execution
- Copy referent of symlinks
- Preserve permission, ownership, etc
- Quiet operation (no output to stdout, stderr)
- Error handling
- Notification of errors
- Detailed logging
- Easy modification of scripts
I'd like to know what everyone else would like in backup software, as when
I'm done I will release it to the community.
I'm currently scripting in Bash, but will probably release in Perl, and
Python as I need practice in all.
- Sebastian
_______________________________________________
RLUG mailing list
[email protected]
http://lists.rlug.org/mailman/listinfo/rlug
_______________________________________________
RLUG mailing list
[email protected]
http://lists.rlug.org/mailman/listinfo/rlug
_______________________________________________
RLUG mailing list
[email protected]
http://lists.rlug.org/mailman/listinfo/rlug
