Networker

Re: [Networker] NDMP Schedling

2009-01-20 11:35:46
Subject: Re: [Networker] NDMP Schedling
From: A Darren Dunham <ddunham AT TAOS DOT COM>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Tue, 20 Jan 2009 16:32:33 +0000
On Tue, Jan 20, 2009 at 10:22:39AM +0000, Davina Treiber wrote:
> A Darren Dunham wrote:
>
>> I'm not sure what you're saying about a level 1 acting as an
>> incremental.  A level 1 and an incremental are different.  In a normal
>> environment, consecutive level 1 jobs will require more volume space
>> than consecutive incrementals.
>
> Not necessarily true. Although you will backup more data with a level 1  
> than an incremental, you will need to keep more volumes to fulfill the  
> retention requirements. If you have a monthly cycle of incrementals you  
> will need to keep up to a month of extra incremental backups, since the  
> incrementals depend on all the other incrementals all the way back to  
> the previous full. With a monthly cycle of level 1 backups the level 1  
> backups will expire at exactly the time when the retention period is  
> reached, depending only on the full.

I'm not suggesting that such a scheme is better or worse than another
overall for all workloads.  But with NDMP, it becomes very difficult to
do true incrs for an extended period of time even if you want to.  I
have one server that I *really* want to.

I don't have any Celerra units, but I've heard that they have a true
incr if asked to do a "level 10" backup.  I don't know if Networker
could ask for that via NDMP or not.  Netapp filers don't have a similar
facility.

-- 
Darren

To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type "signoff networker" in the body of the email. Please write to 
networker-request AT listserv.temple DOT edu if you have any problems with this 
list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER

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