[ale] archiving backups

[Chris Fowler]

On 08/06/2014 02:51 PM, Scott Plante wrote:
> This is an instance where hard links are probably called for--no need to create a dupes directory. Each hard link is equal, and the file only goes away when the last link is deleted.
>
>
> There is a command 'fdupes' that's available in the standard yum/zypper repos (and probably others) that will automatically find duplicate files and, with --link option, create hard or soft links automatically.
>

That looks good.  I think I would want to know what my dupes are. This 
is where symbolic links would help.  I intend to place much of this on a 
drive and then to cloud storage.  I'm not sure how they handle hard links.

This job is starting to suck.

I started going through my old drives and it appears that many of them 
have been wiped.  Another symptom of age.  I've been down the road of 
cleaning my drives.  I just did not throw them out.

One thing I did find was a pack of CVS backups.  I would have done this 
before a software release.  Since these are snapshots of a system still 
in place then I think the best idea is to destroy them and have one 
master CVS backup.   CVS never changed in a reverse fashion.  Same as my 
SVN.

For photos I think I'll get the creation date and store them in 
directories sorted by year/month.  Those would go on the master backup 
and I think also I would put them on 2 32GB USB sticks.  One for 
redundancy.  A question for you guys would be USB stick or SDHC?   With 
the stick I have to worry about more failure than SDHC?  Logical eh?  
Speed is a non-issue for archive.