Networker

Re: [Networker] Another client retention problem

2003-06-06 09:37:13
Subject: Re: [Networker] Another client retention problem
From: Carl Farnsworth <carl.farnsworth AT DIGIDYNE DOT CA>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Fri, 6 Jun 2003 09:37:11 -0400
If you take a look at the retentions of the savesets in your DailySaves
pool, with "mminfo -av -p <volume>", you'll see that your other clients
retentions are indeed what you were expecting - it's the indexes that have
the longer retention, and hold up recycling of these volumes.

This is caused by the fact that NetWorker looks for the longest retention
given for the server's client definition, and uses these for indexes.

If you create a separate Group for the Fullsave copy of the server client,
this should clear up your problem.  (Definition of the Pool may also need
some tweaking).

Regards,
Carl Farnsworth
DigiDyne Inc.

On Thu, 5 Jun 2003 09:48:49 -0600, Craig Ruefenacht
<Craig.Ruefenacht AT US.USANA DOT COM> wrote:

>Hi,
>
>After reading through the recent thread about a client retention problem, I
>noted that I am experiencing a similar problem.  But it doesn't seem that
>the recent thread didn't quite resolve my issue  (or maybe it did but I
>didn't understand how to resolve it).
>
>Here is my situation.
>
>We are running Networker Network Edition 6.1.3 Build 428 on HP-UX 11i,
fully
>patched upto October 2002.  Our clients are a mix of Sun/HP.
>
>I have a group of client machines that have incremental saves performed
each
>night except for Sunday.  On Sunday a full backup is performed.  The setup
>is like this:
>
>        Pool: DailySaves
>        Schedule: incr Mon-Sat, full Sunday
>        Browse Policy: Month
>        Retention Policy: Quarter
>
>So all the backups that are performed are written to the same tape pool.
>One of the Networker clients is the Networker server itself (we have a
bunch
>of BCV volumes that get mounted onto the Networker server for backups).
>
>The above setup has worked fine for us until....  We wanted the full save
of
>the BCV volumes on the Networker server to go to a different pool and have
a
>three year retention policy, while keeping the Monday-Saturday incrementals
>going as shown above.  To do this, in the client setup, I copied the
>Networker Server client entry and changed the copy's settings to this:
>
>        Pool: Offsite
>        Schedule: full Sunday, skip all other days
>        Browse Policy: Quarter
>        Retention Policy: 3 Years
>
>The other (already existing) entry for our Networker server was changed to
>this:
>
>        Pool: DailySaves
>        Schedule: Incr Mon-Sat, skip Sunday
>        Browse Policy: Month
>        Retention Policy: Quarter
>
>So in the client list, there are two entries for our Networker server. One
>is used to perform the daily incrementals (Monday - Saturday) and the
second
>one does the full on Sunday.
>
>After implementing this configuration, the tapes in *BOTH* pools (the
>DailySaves and Offsite) began to have a retention of 3 years, even though
in
>the client setup, for the clients that use the DailySaves pool, the browse
>policy is Month and the retention policy is Quarter. It appears that these
>settings are getting overwritten by the one client (which happens to be the
>Networker server) that is setup to write to the Offsite pool on Sunday and
>the DailySaves pool on all other days.
>
>What we really want is for Networker to honor the browse and retention
>policies that are set in the client properties.  Even though our Networker
>server has two client entries (so that we can write backups from each entry
>to different tape pools and use different schedules), Networker should set
a
>tape's retention policy based on the save sets written to the individual
>tape, not based on a retention policy that's used for a save set which
isn't
>even written to the tape's pool.
>
>Is there any way around this issue?  It doesn't cause us any big problem
>right now because we have a set number of tapes in the DailySaves pool and
>manually recycle the oldest tapes when there is none to write to (tapes
>usually get recycled once every 35 days or so).  Its just annoying to have
a
>retention policy of save sets being written to one tape pool be used on
save
>sets in a different tape pool.
>
>--
>Craig Ruefenacht
>UNIX Systems Administrator
>USANA Health Sciences, Inc.
>http://www.usana.com
>
>
>
>--
>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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

<Prev in Thread] Current Thread [Next in Thread>