Hi All,
I've got a machine that is acting as my backup server which I call
"Valykyrie". Up until this weekend, this machine was running RedHat
7.3 and
Samba 2.2.5. Because of some "political issues", I was forced to
rebuild
the machine with SuSE 8.1 Professional and Samba 2.2.5.
Since the time of the machine coming online with SuSE on Sunday afternoon,
my backup process has been running about 4 hours longer than normal. What I
have happening is, through a script, Valykyrie mounts the administrative
shares of a server's hard drive, copies the data local, compares the data
and then moves on to the next drive or the next server and starts over - as
shown here:
==================================================================================mount
-t smbfs -o
username="domain_name\admin_username",password=<proper-password>
\
//machine-name/sharename$ /mnt/mount-point
cp -a /mnt/mount-point/ /backup
diff -r brief /backup/mount-point /mnt/mount-point
unmount /mnt/mount-point
==================================================================================This
cycle repeats for each required share on each server. <No gripes about
rsync, partial backups, etc. please. This is the way I want it to work.>
This amounts to about 45GB of data a night. This used to take 8 hours
total, now it's taking nearly 12. Nothing changed hardware-wise or in my
script, only the distro.
Could just the change to SuSE cause this much of a performance loss? Is
there anything I can do to increase the performance? I'm not looking for
huge improvements, but I need these additional 4 hours back so that my users
aren't working on data while the backup is taking place.
Thanks,
Kevin L. Collins, MCSE
Systems Manager
Nesbitt Engineering, Inc.
(859) 233-3111 x24
