BackupPC-users

Re: [BackupPC-users] Status of fuse for viewing backuppc backups

2008-12-16 13:02:24
Subject: Re: [BackupPC-users] Status of fuse for viewing backuppc backups
From: Les Mikesell <les AT futuresource DOT com>
To: "General list for user discussion, questions and support" <backuppc-users AT lists.sourceforge DOT net>
Date: Tue, 16 Dec 2008 11:59:46 -0600
Jeffrey J. Kosowsky wrote:
>
>  > But the thing I alway thought should happen was some sort of merge with 
>  > bacula code.  That is, either make backuppc able to talk to bacula's 
>  > client agent, or put backuppc's file pooling code into bacula's disk 
>  > archive handling.
> 
> I agree - I think the ideal combo would be:
>   - Pooling & compression like with BackupPC (though I would add full
>       md5sum checksums which you would get anyway I think if we upgrade
>       to protocol 30)

Md5sum might be a natural along with rsync protocol 30 but for anything 
else it is going to force reading the whole file to compute before 
attempting to link, make the directory operations slower due to the 
size, and you still have to deal with possible collisions.

>   - Database backend like Bacula (this would get rid of the platform
>       dependence of hard links along with all the associated limitation
>       and backup issues caused by them; plus, it would make it much
>       easier to extend to allow backup of acls, extended attributes, and
>       any other filesystem entry needed to accurately recreate the
>       backup)

I doubt if a database is a better place to store files than a filesystem 
- or if it can do the needed atomic operations any faster.

>   - Choice between simpler server+rsync method like BackupPC (for SOHO
>       or personal use) vs. more sophisticated client/server approach of
>       Bacula

While it is nice to not specifically require a client, it doesn't add 
much extra work since you have to at least set up ssh keys anyway.

>  > As for fuse, it is a nice idea but I doubt if you'll like the 
>  > performance on operations that have to uncompress the files.
> 
> It is fine for small numbers of files and poking around. Certainly
> faster than navigating the CGI interface ;)

But if you have small numbers of files and a need to poke around 
different versions you should probably be using a version control system 
to manage them anyway (subversion, git, etc.).  In fact there is a 
fuse/webdav layer for subversion somewhere

-- 
   Les Mikesell
     lesmikesell AT gmail DOT com


------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you.  Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
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/