Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 03 May 2010 15:19:03 +0200 (CEST)
From:      sthaug@nethelp.no
To:        current@freebsd.org
Subject:   Re: SUJ update
Message-ID:  <20100503.151903.74740368.sthaug@nethelp.no>
In-Reply-To: <20100503140438.262539xlm87yp0ao@webmail.leidinger.net>
References:  <4BDCE05A.5020307@FreeBSD.org> <20100502.073857.74726756.sthaug@nethelp.no> <20100503140438.262539xlm87yp0ao@webmail.leidinger.net>

next in thread | previous in thread | raw e-mail | index | archive | help
> >> I would vote for decoupling. If I have SU on, then enable journaling,
> >> then disable journaling, I would expect SU to still be on.
> >
> > Fully agreed. I see no reason why these sould be coupled.
> 
> It does not look like it is a prerequisite to have SU enabled when you  
> want to enable SUJ. So I assume SUJ implies SU, and as such I think  
> you can agree that it is not easy to determine at disable time of SUJ,  
> if the FS was SU before or not.

If SUJ requires SU then IMHO tunefs should prohibit setting SUJ unless
SU was already enabled, with a nice explanatory error message if needed.

Looking at it from a slightly different angle - assume I have a file
system with SU enabled, and I want to experiment with SUJ. So I enable
SUJ. When I'm finished testing, maybe I want to disable SUJ again. I
would be *highly surprised* (badly breaking POLA) if SU was disabled
at the same time.

> So whatever the consensus is (disabling SUJ does or dosn't enable SU),  
> the man page needs to tell what it does.

Agreed.

Steinar Haug, Nethelp consulting, sthaug@nethelp.no



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100503.151903.74740368.sthaug>