Brian,
In general it's best (performance wise) to let NetWorker figure out how many
drives to use rather than trying to control it via group settings. If you just
want to make all the drives go, then it would be simpler to set your target
sessions per device to 1 and let NetWorker do the load balancing. A lot of
this depends on the number of clients and data streams you are backing up as
well as your parallelism and target session settings.
Things to look at would be how is your NW server parallelism currently set?
What is the target session setting for each tape drive? How many clients are
you backing up at one time? How many data streams are you getting from each
client during the backup? Is the data throughput speed you are seeing during
backups, near the maximum that your drives are able to support? Do your client
definitions limit the number of data streams a given client can produce at one
time? (Save sets and client parallelism settings?)
In general, you could monitor the through put on each tape drive during the
backups with your current settings and make a note of them. Then experiment
with changing the server parallelism and then later, the target sessions per
device to see if it improves your throughput and reduces your backup time.
This would be a more accurate way to determine whether or not your drives are
being underutilized. I've seen cases where forcing NW to use more drives
actually hurt performance and made backups run longer.
Joseph Ortiz
DSMCS, Inc.
----- Original Message -----
From: "Spindler, Brian" <John.Spindler AT RUESCH DOT COM>
Date: Tuesday, February 24, 2004 1:48 pm
Subject: Re: [Networker] Tape utilization
> Thank you Anuj. I will check the properties on my Groups.
>
>
> Thanks,
> Brian
>
> -----Original Message-----
> From: Anuj Mediratta [anuj AT ACE-DATA DOT NET]
> Sent: Tuesday, February 24, 2004 2:02 PM
> To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
> Subject: Re: [Networker] Tape utilization
>
> Hi Brian,
>
> If you have 4 groups to run simultaneously, you can select a specific
> drive for each group while configuring the group properties.
>
> Regards,
> Anuj
>
> On Tue, 24 Feb 2004 11:32:35 -0500, "Spindler, Brian"
> <John.Spindler AT RUESCH DOT COM> wrote :
>
> > Legato 7.1
> >
> > I have 4 tape drives and 9 times out of 10 only two of them are
> being
> > utilized. How can I make it so that all four will be used
> during
> > backups? Or how can I just make sure that whatever job it's
> doing only
>
> > needs two drives?
> >
> >
> > Thanks,
> > Brian
> >
> > -----Original Message-----
> > From: Dan Tulovsky [dtulovsky AT sbigroup DOT com]
> > Sent: Tuesday, February 24, 2004 11:07 AM
> > To: Legato NetWorker discussion; Spindler, Brian
> > Subject: RE: [Networker] Unload retry
> >
> > I get this once in a while too. If it eventually manages to
> unload
> > the tape, then it's ok. It means networker is waiting for the tape
> drive...
> > If it gets stuck, it might be a hardware issue.
> >
> > Dan
> >
> >
> > -----Original Message-----
> > From: Legato NetWorker discussion
> > [NETWORKER AT LISTMAIL.TEMPLE DOT EDU]
> > On Behalf Of Spindler, Brian
> > Sent: Tuesday, February 24, 2004 10:44 AM
> > To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
> > Subject: [Networker] Unload retry
> >
> > Hi guys/gals,
> > I keep getting this message (Legato 7.1):
> > NetWorker media: (info) unload retry for jukebox `QualStar8466'
> failed
> > - will retry again.
> >
> > I am very new to Legato, how can I fix this?
> >
> >
> > Thanks,
> > Brian
> > --
> > 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.
> > =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
> >
> > --
> > 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.
> > =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
> >
> >
> >
>
> --
> 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.
> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
>
> --
> 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.
> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
>
--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|