im not too clear on this so i want to ask for a bit more clarity
if i have 10 clients and each client has a file called file1.jpg and all 10
of these files are absolutely identical, does backuppc have just 1 copy of
this file in the pool? what if the files are exactly the same exept the
timestamp is different?
from what i know, if the timestamp is different, then the file would be
considered different. could an MD5 be made of each file that has an
identical name and size and then compare the MD5 and pool that file
appropriately?
On Dec 14, 2007 12:16 PM, Les Mikesell <[EMAIL PROTECTED]> wrote:
> Mark Best wrote:
> > I have some questions about pooling and rsync checksum.
>
> They happen separately, and may have different answers.
>
> > Let's suppose I have two servers Server1 and Server2.
> >
> > 1) For redundancy and performance reasons I have the two servers syncing
> > data between each other.
> >
> > 2) Server1's file structure is: C:\server1\backup.tiff and it syncs
> > using DFS to Server2's file: C:\server2\backup.tiff.
> >
> > 3) The files are technically the same file – but are on different
> > servers and have different directories they reside in.
> >
> > 4) Server1 is scheduled to backup at 6:00pm Server2 is scheduled at
> 5:00am.
> >
> >
> > During a full rsync backup will the rsync client transmit both files?
> > (Even if they are in different directories.)
>
> Yes, the rsync comparison is done against the previous copy in the
> per-pc tree, so whether the file (or differences) are transmitted will
> depend on the specific machine contents (current and previous).
>
> > If two files are transmitted will BackupPC pool both of these like-files
> > into one?
>
> Yes, after the complete new copy has been reconstructed, the pooling
> mechanism will match it up with an existing pooled duplicate.
>
> > Based on the above how will rysnc's checksum check the files above
> > during an incremental backup?
>
> It's the same as a full, except that if the file name, timestamp, and
> length match the existing copy it skips the block-checksum comparisons.
>
> > If a change is made to Backup.tiff, and Server1's incremental backup
> > runs first and 'updates' the file to the pool – will Server2's rsync
> > 'Passover' the file because it's been updated by Server1's rsync
> > transmittal?
>
> No, the transmission is controlled by the per-pc contents.
>
> > Basically, I'm interested in maximize pooling efficiency and shortening
> > backup times – but worry because of the differences in directory folders
> > and NTFS/DFS file system and whether the Inode/meta-data pooling is
> > forgiving enough.
>
> If you have a 'master' copy and a mechanism to clone it exactly to one
> or more duplicates, why not exclude the copies that you can already
> reproduce and just back up the master?
>
> --
> Les Mikesell
> [EMAIL PROTECTED]
>
> -------------------------------------------------------------------------
> SF.Net email is sponsored by:
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services
> for just about anything Open Source.
>
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> BackupPC-users mailing list
> [email protected]
> List: https://lists.sourceforge.net/lists/listinfo/backuppc-users
> Wiki: http://backuppc.wiki.sourceforge.net
> Project: http://backuppc.sourceforge.net/
>
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
BackupPC-users mailing list
[email protected]
List: https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki: http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/
