xfs
[Top] [All Lists]

Re: TAKE - mount msgs

To: Ethan Benson <erbenson@xxxxxxxxxx>, linux-xfs@xxxxxxxxxxx
Subject: Re: TAKE - mount msgs
From: Seth Mos <knuffie@xxxxxxxxx>
Date: Tue, 04 Sep 2001 15:23:12 +0200
In-reply-to: <20010904043217.U14519@xxxxxxxxxxxxxxx>
References: <20010904225323.A344731@xxxxxxxxxxxxxxxxxxxxxxxx> <200109041139.VAA08035@xxxxxxxxxxxxxxxxxxxxxxx> <20010904134437.A28205@xxxxxxxxxxxxxxxxxxx> <20010904225323.A344731@xxxxxxxxxxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
At 04:32 4-9-2001 -0800, Ethan Benson wrote:
On Tue, Sep 04, 2001 at 10:53:23PM +1100, Nathan Scott wrote:
> hi Andi,
>=20
> On Tue, Sep 04, 2001 at 01:44:37PM +0200, Andi Kleen wrote:
> > On Tue, Sep 04, 2001 at 09:39:42PM +1000, Nathan Scott wrote:
> > > Consensus seems to be less chatty on mount - this should do the trick.
> > > Can still get at these messages with a debug build (as is the case for
> > > the nightly QA runs) where these messages are still quite useful.
> >=20
> > I think it would be better to keep them; as it makes debugging even on
> > production machines easier.
> >=20
>=20
> _Now_ you pipe up!  Well, I don't think the "finished clean mount"
> message is particularly useful, but I'll put the initial message
> back in at the start of a mount, I think, so that in the normal case
> its just one message per mount.

please don't, or make it a compile time option.  it makes an ugly mess
of boot messages and clutters the logs with really useless data.

> If anyone else really wants the message at the end of a clean mount,
> I guess I can put that back too...

make it a compile time option if some people want it. =20

CONFIG_VERBOSE_MOUNT or something. =20

no news is good news.

I like having at least one message that it mounted a certain fs. I can then use dmesg to see if it actually worked, and if it needs to recover or if it fails I like to see it as well.

One line is enough. What I don't know is if you could stick these messages on one line. So the first mounting message without a \n and the succes or fail message after it.

Don't know if it lets us do that.

Cheers

--
Seth
Every program has two purposes one for which
it was written and another for which it wasn't
I use the last kind.


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