My original question:
I need to make a decision quickly about whether to buy the Legato
Networker file backup software for campus use here - we have about 200
ethernetted Suns, as well as other manufacturers' equipment. I'd very
grateful if you'd send me your comments about it, as I'd like to know
if there are any problems with its use. Please reply direct to me and
I'll send a summary to the list. Thank you.
Summary:
Most people who responded liked the product. Most people also said
that the indexes occupied a lot of space, a typical figure being 10%
of the size of the backed-up disk space for the indexes.
One person said that he is using Legato Networker with a network
including Sun 3s, Sun 4s, Auspex, Solbournes and HPs, and has had no
problems. He had 9 master servers backing up 3-5 slave servers each,
the latter having 2-15 GB of disk space. He said that it took a very
long time do backups for fileservers with more than 6GB of disk. He
had successfully recovered many filesystems. He recommended the
product.
People liked its ability to do unattended backups, through switching
automatically to the next tape drive when the current drive's tape is
full.
Some people didn't like the fact that it used its own format, rather
than rdump, but also said that it's faster than rdump and that the
index makes restores a lot easier.
One person commented that at present Legato only support writing of
up to 2 GB on Exabyte 8500s, although the tapes can hold 5GB.
Another commented that Networker changes the last use time of files
that it backs up, making it hard to find files that have been left
around for months without being used. The same person commented that
there was no easy way to find out if disk space for the index is full,
since empty slots are normally re-used.
A couple of people said that it could generate a lot of network
traffic.
Thanks:
Many thanks to the following for their helpful contributions:
Mark Harrow <mdh@com.amdahl.da>
cdr%kpc.com%kpc @ net.uu.uunet
Oran Davis <oran@com.amdahl.spg>
Julian Dryden <dryden@au.oz.su.physics>
Greg Morrison <morrison@ca.carleton.ccs>
Stergios <stergios@edu.stanford>
esanborn%cadence @ net.uu.uunet
Peter Smith <peter@edu.stanford.sucia>
Jim Seavey <jwseave@com.pacbell.srv>
Edward Chien <edward%tekbspa@net.uu.uunet>
Brian Kelley <brian%euler%dgsi%hela%caen@sharkey.cc.umich.edu>
stan @ nli.com
Matt Cohen <sysnmc%magic706%chron@uunet.uu.net>
fritchie @ acc.stolaf.edu
----------------------------------------------------------------------
Ian Doak, Computing Laboratory, Newcastle University, NE1 7RU, England
email: I.D.Doak@newcastle.ac.uk
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:16 CDT