[rdiff-backup-users] Is it normal for the first time backup to be SSLLLOOOW

[George Gallen]

If I did a straight cp -Rp /foo /bar it would have taken maybe 30-35mins
So far rdiff-backup has been running for over 3 hours (for one filesystem).

The from and to are on the same machine, however, they are on
different drives. The from drive is a Raid array of 10k rpm SCSI's and the
to drive is IDE 7200rpm.

I'm assuming it's taking longer for the initial build of indexes and such.
or am I assuming incorrectly? At this rate the rdiff-backup will take apx
15 hours to run completely (for the 5 filesystems). I'm substituting this for
a direct backup to tape (which takes 2-3 hrs), then I'll backup just the
mirror at night (excluding the differences directory). I'm trying to setup a
system to keep a range of past backups available online.

I'm hope that subsequent runs will be MUCH MUCH shorter (for the incremental).

I'm currently running RH 7.2 (can't upgrade just yet, application dependent) but
python and libs are most current. It's running on a dual 3ghz Xeon Processor
system, with a fairly low amount of usage.

Is this normal?
Thanks

George

George Gallen
Senior Programmer/Analyst
Accounting/Data Division
address@hidden
ph:856.848.1000 Ext 220

SLACK Incorporated - An innovative information, education and management company
http://www.slackinc.com