On 29 Jan, Jamie Wilkinson wrote:
>  So what happens is you say "snapshot volume 3 and use the remaining space 
>  as a snapshot volume", at which point you get a new snapshot block 
>  device which points to volume 3, and your old block device which you 
>  thought you had mounted on /home is now volume 4 (which is some made up 
>  number which refers to a combination of volume 3 and the unallocated space) 

Maybe I'm missing something, but all this discussion of using tapes or
external drives that you swap between, still suffer from the original
problem I was asking about: when you write back onto the same area you
erase your old backup.

It doesn't even matter whether it's an incremental or full backup.

The vulnerability is in how quickly you reuse media, and I was simply
observing that just cycling between media gives you a very short window
to realise you lost data, and to then restore it from backup.

That's all.

So yes, if you have storage capacity to store 100 complete full
backups, on separate media, then you have 100 days maximum to notice
the data loss and restore it.

But that seems neither an elegant nor an efficient use of resources.

I think my new scheme, which DaZZa's remark about "grandfather Father Son"
method triggered, is moderately elegant and efficient.  I'm sure a good
pure mathematician could easily write down an expression for an optimal
method that maximised the window of recovery period for a given number
of backup media.  Mine is just a stab in that direction.

(6 dailies using 1 media, 4 weeklies using 1 media, full backups using
N media, where N is noticeably bigger than 1 :-), and full backups
triggered by overflowing 1 media.)

luke

-- 
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html

Reply via email to