BackupPC-users

Re: [BackupPC-users] BackupPC 3.x Pool on BTRFS?

2016-02-15 15:21:22
Subject: Re: [BackupPC-users] BackupPC 3.x Pool on BTRFS?
From: Carl Wilhelm Soderstrom <chrome AT real-time DOT com>
To: cvoelker AT knebb DOT de, "General list for user discussion, questions and support" <backuppc-users AT lists.sourceforge DOT net>
Date: Mon, 15 Feb 2016 15:20:25 -0500
On 02/14 11:01 , Christian Völker wrote:
> I did find only a single reference to the topic- but already some years ago.

Look for discussions of BackupPC on ZFS. It's all the same issues.

> Deduplication feature:
> BackupPC hadrlinks identical files to save storage space which is
> already deduplication. But it uses hardlinks to do so- which frequently
> causes issues as you can see in this mailing list. So possibly diasble
> hardlinking in BackupPC (if possible) and let BTRFS do the work with
> deduplication? Disadvantage for BTRFS: it is only a "out-of-band"
> deduplication- so you have to perform the dupe detection by cron or so.

The deduplication is built into the storage system of v3 and below, so it
can't be disabled. v4 does deduplication differently, I do not know if it
works better with BTRFS/ZFS.
 
> Online-Compression:
> BTRFS uses online-compression- I am unsure if there would be an
> advantage if compression is done by filesystem. At least IMHO it would
> be easier to restore directly from the pool instead of cpool through bash?

I believe this has been discussed before, and I think the answer is "don't
try it" but I could be wrong.

> Pool-migration/ move:
> If you decide to move your pool to a different location you are happy if
> you have it on something like LVM (or completely virtualized). Then you
> can move the underlying device. Disadvantage: LVM and other tools move
> the full device. If the pool is just filled up to 70% it even transfers
> the 30% of empty space. With btrfs receive and btrfs send the filesystem
> will only transfer the used blocks.

I believe some people do this with ZFS.

> COW-Feature:
> With copy-on-write btrfs writes changed block on a different location
> and then refers to this new location. Has advantages for snapshots and
> so on. But disadvantages for larger files (like virtual disk images) as
> they spread among all devices which causes lots of seeks on rotating
> devices like HDDs. How about pool? Am I right this does not matter for
> backupPC as it always writes full files only? Even rsync-based backups?

I'm not sure how this relates. I do not think I understand the question
fully.

-- 
Carl Soderstrom
Systems Administrator
Real-Time Enterprises
www.real-time.com

------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
BackupPC-users mailing list
BackupPC-users AT lists.sourceforge DOT net
List:    https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:    http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/

<Prev in Thread] Current Thread [Next in Thread>