[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-tar] Question on the one-filesystem option
From: |
Gene Heskett |
Subject: |
Re: [Bug-tar] Question on the one-filesystem option |
Date: |
Mon, 28 Jul 2014 04:38:32 -0400 |
On Sunday 27 July 2014 17:29:49 Alan Hodgson did opine
And Gene did reply:
> On Friday, July 25, 2014 06:00:35 PM you wrote:
> > No, at the end of the day, we are on the same page Deb. Its a matter
> > of scale of course and that cannot be helped. I just want my
> > softlinks to actually be backed up, without any dereference, and its
> > not, even with tar 1.27.1, happening.
>
> You can test tar easily enough. --one-file-system does not skip or
> dereference symlinks. Tested with 1.23 and 1.27.1.
>
> I did some other tests and with Red Hat's tar 1.23 (patched for xattrs
> support), amrecover does restore symlinks, regardless of which file
> system is targeted. This with amanda 3.3.0.
I took the --one-file-system option out of the options that amgtar passes
to tar, and at least the verbose error messages are now gone. So I ran
amrecover Daily, and checked 2 of the 4 missing listings, and they now are
there in the backup database, I was able to chase one of them in /var/lib
2 more directory levels deeper than the failure point now, so I believe
the problem to be solved in a usable manner.
However, in one failure path it quoted, there are no softlinks. It simply
did not like /var/lib/nfs/rpc_pipefs, which actually goes 2 or 3 levels
deeper, encountering no softlinks as it goes.
So I'm still standing here on my stump and proclaiming that tars "--one-
file-system" implementation is broken. And has been since somewhere post
1.22.
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
- Re: [Bug-tar] Question on the one-filesystem option,
Gene Heskett <=