Re: Reducing noise in dmesg output

On Wed, Sep 2, 2009 at 8:39 PM, James Butler <sweetnavelorange_at_gmail.com>wrote:

> 2009/9/3 Nick Hibma <nick_at_van-laarhoven.org>:
> >> What is irrelevant is subjective...  I mean does the average user really
> >> need to see anything in dmesg?  Please don't change agp_i810.c.  A
> >> verbose boot is incredibly noisy and rarely needed for debugging
> >> anything except the most deep rooted of issues.
> >
> > Please state arguments instead of this 'I think it is useful'. That's not
> > going to cut it. If you feel strongly about the info that is being
> produced,
> > an option would be to produce a (tested!) patch that combines more
> > information on one line as a compromise.
> >
> > FreeBSD has historically been producing very limited output on dmesg.
> Linux
> > is very noisy (ever noticed the copyright notices right in the middle of
> > your list of PCI devices?). Even they have decided that they should hide
> > this behind coloured 'ok/failed' texts in some distributions.
>
> This seems like an important distinction - the information which needs
> to be available with dmesg and the information best shown to the user
> at startup are not necessarily the same. The hypothetical "average
> user" probably wouldn't care if there were *no* kernel messages shown
> on startup. The Xubuntu box I'm writing this from shows only GRUB
> messages before the login prompt on tty1, and only service startup
> messages on tty8 (not to suggest that FreeBSD ought to be more like
> Ubuntu...).
>
> IOW, why bother painting it at all? Plain galvanized steel is very
> durable, and quite attractive when kept clean ;-)
>
> -James Butler
>

Doesn't the 'boot_mute' (loader) environment variable do just that?

>From the man page:

     boot_mute
               All console output is suppressed when console is muted.  In a
               running system, the state of console muting can be
manipulated
               by the conscontrol(8) utility.

- Justin