Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 10 Dec 2000 13:26:38 -0400
From:      "Jeroen C. van Gelderen" <jeroen@vangelderen.org>
To:        Poul-Henning Kamp <phk@critter.freebsd.dk>
Cc:        Julian Elischer <julian@elischer.org>, arch@FreeBSD.ORG
Subject:   Re: inheriting the "nodump" flag ?
Message-ID:  <3A33BCCE.844B35B4@vangelderen.org>
References:  <97668.976451080@critter>

next in thread | previous in thread | raw e-mail | index | archive | help
Poul-Henning Kamp wrote:
> 
> In message <3A336D68.2B454829@elischer.org>, Julian Elischer writes:
> 
> >> I would like to propose that directories and files inherit the
> >> nodump flag if it is set on the directory they are created in.
> >>
> >> Comments ? protests ?
> >
> >I thought that it was supposed to stop the tree walker from walking
> >down through the tree beyond that point if it is on a directory..
> >
> >in other words, if you set it on a directory it is supposed to
> >'prune' the entire tree.. sounds like a problem with the
> >tree walker rather than the kernel.
> 
> Well, dump's tree walker is rather peculiar, so I don't blame
> anyone for the way it is implemented currently.

This answer does not address the real point Julian is trying 
to make. Trying to fix a buggy dump implementation by patching 
the kernel is the wrong approach. It doesn't matter how peculiar 
the tree walker is, if it's buggy the fixing needs to be done 
there.

It looks like NetBSD have already addressed the problem a
year ago:

 http://lists.openresources.com/NetBSD/tech-kern/msg00453.html

for the beginning of a thread discussing the problem. This
was the first hit Google returned for "nodump flag". Basic 
research is cheap these days...

A look at the NetBSD PR in question (6705) reveals:
http://www.NetBSD.org/cgi-bin/query-pr-single.pl?number=6705
[...]
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->closed 
State-Changed-By: bouyer 
State-Changed-When: Tue Mar 9 09:32:08 PST 1999 
State-Changed-Why:  
Functionality added [to dump I presume], but differently. 
^^^^^^^^^^^^^^^^^^^

It seems that fixing dump is the correct approach. Changing
the semantics to nodump almost certainly is not as Julian 
pointed out in his other mail (<3A336EA3.9F1FE318@elischer.org>). 

I can now even add another reason why changing the kernel isn't
such a good plan: inter-BSD compatibility.

Cheers,
Jeroen
-- 
Jeroen C. van Gelderen          o      _     _         _
jeroen@vangelderen.org  _o     /\_   _ \\o  (_)\__/o  (_)
                      _< \_   _>(_) (_)/<_    \_| \   _|/' \/
                     (_)>(_) (_)        (_)   (_)    (_)'  _\o_


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3A33BCCE.844B35B4>