Networker

Re: [Networker] Possible problem with directive

2003-08-28 13:31:55
Subject: Re: [Networker] Possible problem with directive
From: George Sinclair <George.Sinclair AT NOAA DOT GOV>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Thu, 28 Aug 2003 13:31:48 -0400
That's an excellent point! The browse period is one month. I ran
nwrecover again using the default browse time of today and checked the
times on the affected directories, and they're older times and dates
from the last full since they've not changed since then. I guess I'm
confused here, though, because I thought as soon as you placed a 'skip'
on something you'd no longer see it in the nwrecover window unless you
browsed back to the last time it was not being skipped or earlier?
Doesn't 'skip' force NetWorker to remove it from the listing unless you
browse back to the time before you created the skip?

THe other strange thingh, though, is that while all of these are listed
in the recover window, not all of them show up when I check file
details. Some do, some don't.

George

Charles Heynig wrote:
>
> You didn't mention the backup times on the directories listed.  Is
> it possible that they are just showing up because they are still in
> their browsable time period.
>
> Charles Heynig
> Data Storage Specialist
> Western Michigan University
> 269-387-6200
> >>> George Sinclair <George.Sinclair AT noaa DOT gov> 08/28/03 12:05 PM >>>
> Hi,
>
> Having a problem with a directive. I'm afraid it may not be working.
> It's a little confusing, so please bear with me. First, we're
> running
> 6.1.1.build38 on both storage node and primary server. Primary
> server is
> running Solaris 8 and storage node is Linux RedHat 7.3. I have two
> client instances. The first uses the standard unix directive and now
> specifies 11 savesets: /pathname/olddir1-olddir5,
> /pathname/data/olddir6-9 and olddir13-14.
> NOTE: olddir10-12 and olddir15 were removed from the saveset listing
> as
> described below.
>
> I'm just using the 'olddir' naming convention for the purpose of
> this
> posting. The real directories do not have consistent naming
> conventions.
> The second instance backs up 'All' and uses a custom directive which
> until recently has looked like this:
>
> << /pathname >>
> null: olddir1
> null: olddir2
> null: olddir3
> null: olddir4
> null: olddir5
>
> << /pathname/data >>
> null: olddir6
> null: olddir7
> null: olddir8
> null: olddir9
> null: olddir10
> null: olddir11
> null: olddir12
> null: olddir13
> null: olddir14
> null: olddir15
>
> Recently, I decided to get rid of directory: olddir15, and I no
> longer
> wanted to back up the 3 directories: olddir10-12, so I removed these
> 4
> pathnames from the list of savesets for the first client instance,
> removed the directive entry for olddir15 and changed the 'null'
> entry to
> a 'skip' for the other 3 in the directive for client instance 2
> (shown
> above).  Then I decided I wanted to also skip 4 directories that
> were
> not previously listed among the savesets for instance 1. These were
> being captured in the second client instance that runs 'All', but
> were
> never listed in the directive, so I added these with 'null'. The
> directive now looks like this:
>
> << /pathname >>
> null: olddir1
> null: olddir2
> null: olddir3
> null: olddir4
> null: olddir5
>
> << /pathname/data >>
> null: olddir6
> null: olddir7
> null: olddir8
> null: olddir9
> skip: olddir10
> skip: olddir11
> skip: olddir12
> null: olddir13
> null: olddir14
> skip: newdir16
> skip: newdir17
> skip: newdir18
> skip: newdir19
>
> What I noticed was that when I run nwrecover from the client, I
> still
> see all the olddirs showing up! Why would the ones that are being
> skipped show up? I set the browse date to the most recent backup
> (and
> incremental that ran last night), and they're still listed. None of
> these directories has changed in several weeks. Maybe I should have
> all
> the 'null' entries listed first? Maybe my directive is tool long?
> Maybe
> use .nsr file on client instead? Any explanations? It's not like
> client
> instance 1 is coming along and adding them back into the index
> because
> they're not listed for instance 1.
>
> Any help would be appreciated. We're running fulls this weekend, and
> I
> don't want these guys getting backed up.
>
> 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.
> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
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>