Re: [Networker] How much time do you spend on backups?



George Sinclair <George.Sinclair AT noaa DOT gov>
Sent by: Legato NetWorker discussion <NETWORKER AT LISTMAIL.TEMPLE DOT EDU>
31/10/2003 07:01 AM
Please respond to george.sinclair


        To:     NETWORKER AT LISTMAIL.TEMPLE DOT EDU
        cc:
        Subject:        [Networker] How much time do you spend on
backups?


Just curious how much time some of you spend on your backups. Clearly,
when new equipment is being installed or major software updates, you're
gonna be spending a lot of time, but let's just say on average:

Environment:
Server:   Dell 1300 / RH Linux 8.x / Networker 7.0
Library:  Qualstar with 3 AIT-3 drives and 126 slots
Clients:  58
Total Data backed up:   7 TB

Methodology:   Basically a quarterly system, with pools for
       incremental, level 5/7, full, and index.   (Full backups only
    once a quarter)

1. How much of your work week is spent on backups.

The time varies enormously -- can be as little as 10 minutes (checking
the log files for successful completions) or as much as 10 hours (system
hangs when labeling a tape / time to get ready for full backup and need
to
remove current tapes to get ready)


2. How many people are involved.

Just me.


3. If you are spending a lot of time, what's your justification? How do
you justify it?

The problem here is that the time oscillates quite a bit.   I would say
that backup
has weirdnessnesses more than most applications and most of them wind
up being
SCSI or timing related.   The majority of the time, a person supporting
Networker would have nothing to do.   I cram it into other pieces of my
day
and other people understand that it's high-priority and might interrupt
other things I'm doing.   I probably should pass it off to one of my
staff members
but I'm comfortable with it and very concerned that it work as I want
it to.

I can imagine that a big Networker server failure could cause a huge
amount of time to be
needed, but we can't have someone hired to wait for this occurrence.

The number of small interventions when tapes haven't mounted, or a
drive just doesn't rewind
is quite high.   We get around much of this by having large-ish RAID5
file servers that
do snapshotting -- we have the snapshots remain for a week (and could
do more if necessary).
That way, even if our Backup server was having major issues, we have a
week's grace period in
case someone deletes a file.   We encourage people to use our servers
rather than
keeping data locally.

If you're going to do it our way, the person in charge must have a
flexible
schedule, be comfortable with Unix and its command line interface.   My
impression is
that this would be harder if the server was an NT-based one.

Also, get the biggest library you can and keep plenty of tapes around.
 Then,
do ask yourself if the incremental backups couldn't go to a raid array.
  It's the
tapes that are the time-consuming part of Networker and any backup
system.

                                                                                
       Matt Temple


Backups at my shop involve just myself and one tape operator. That's
it,
just the two of us. Personally, I find that this whole backup thing is
pretty much a full time job unto itself. There is always plenty of work
to be done, and I can only automate so much, but some folks seem to
think otherwise. They say that with proper automation, a backup
specialist's job can be reduced by 80%, and if not, well then things
just must not be configured or set up right. I tend to doubt that. Can
anyone help me to justify that backups are not some simple thing that
you just quietly usher away at night, only to have some magical reports
and scripts just do everything for you? Would like to hear it.

Thanks.

George

--
Note: To sign off this list, send a "signoff networker" command via
email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=





This message is intended for the addressee named and may contain
confidential information. If you are not the intended recipient or
received it in error, please delete the message and notify sender.
Views
expressed are those of the individual sender and are not necessarily
the
views of their organisation.

--
Note: To sign off this list, send a "signoff networker" command via
email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

========================================================
Matthew Temple                Tel:    617/632-2597
Director, Research Computing  Fax:    617/632-4012
Dana-Farber Cancer Inst       mht AT research.dfci.harvard DOT edu
44 Binney Street,  M L105     http://research.dfci.harvard.edu
Boston, MA 02115              Choice is the Choice!

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=