Why is my backup smaller than the original HD?

[uelef]

Good morning - here I am again...
The reason why I bought the Western Digital drive was that it was one of the fastest and one of the quietest. The internal Seagate drive that was delivered with G5 is much louder when it reads and writes data. I don't like noisy HDs.

I will make a little backup session today to see what's going on with the startup time. I will backup my Seagate drive data to an external FW drive then clone my Western Digital boot drive to the Seagate drive and have a look what happens to the startup time using another internal HD for booting Mac OS X.

Regarding the amount of data on my Macintosh HD and my FW clone: I found on my Macintosh HD on the hidden Volumes folder a folder called "LaCie Sicherung" (means "backup") with a Applications folder in it that has about 23 GB. I guess this folder is responsible for the difference of data on both drives. This folder maybe is the rest of a unfinished backup.

Ulf
________
marijuana bubbler

[dnanian]

Did you use Deja Vu at some point?

[uelef]

No, until now I did not know Deja Vu. But what I used was CarbonCopyCloner - but it did not work with 10.4 and 10.4.1 - now with 10.4.2 it works again. But cloning a drive with CCC takes a bit longer than with SuperDuper!

For what problem should Deja Vu maybe be responsible?

Ulf
________
FES125 Pantheon

[dnanian]

Well: Deja Vu and CCC use psync to do scheduled backups. But, at least in the case of Deja Vu, certain versions of it did not check to see if the drive was mounted before actually starting the copy.

As such, they'd copy to the mount point, in /Volumes, and if the drive wasn't there, the contents of the backup would be copied into a folder rather than to the drive (since both basically look the same).

This can also happen when a drive fails -- drops off the bus -- during a backup operation. The OS might not indicate a problem has occurred quickly enough, and then the mount point which -- again -- looks like a folder, would get created like any other.

The end result is the same, regardless: your data ends up in a folder in /Volumes, rather than on the actual drive you thought it was being copied to.

I hope that helps to explain...

[uelef]

Dave, thanks for your explanations. Does SuperDuper! not use psync?

I today made my announced backup session - and now my Tiger boot disk is back on the Seagate drive. And really the startup time now is much shorter. It really had to do with the Western Digital harddisk. Strange! All hardware tests say it is ok but it's slowing down my login time.
Could there be something wrong with the jumpers? I guess I cannot exchange the drive on warranty because there's not really something wrong. But on the other side I do not want to use a HD that causes these problems...

You wrote about the Maxtor Max Line III. No doubt, they are fast, but are they also quiet? I don't want the HD to be the loudest component in my G5... I hate these loud r/w noises of HDs (I own a LaCie FW drive that is that noisy - for someone who's making music with his G5, noisy HDs are not acceptable!).

Ulf
________
500E

[dnanian]

We do not use psync: our cloner is custom built, from the ground up, for speed and accuracy. Try benchmarking it against psync.

SATA drives don't really have jumpers, as far as I know... so I don't think it's jumpers. I just think it's a slow drive.

I think the MaxLine III is (very) quiet, but you'd have to check it yourself to be sure. I've also heard the most recent Seagates are much quieter than the older ones, too.

[uelef]

Well, I can confirm that SuperDuper! works faster than CarbonCopyCloner - although I did not benchmark it.

When I use Xbench that also checks disk speed I cannot see any difference between my both internal drives: the Western Digital and the Seagate. That's why I cannot explain that booting with the WD drive takes so much longer than with the Seagate drive. There must something wrong with the WD drive - but I don't know what.
________
Enzo Ferrari (car) history

[dnanian]

Well, XBench clearly isn't giving you the real-world benchmark you're seeing on your own machine! Another reason benchmarks aren't always that useful...