Re: Noisy ipmi0 after bootup? (really other kernel noise)

[Paul Goyette <paul%whooppee.com@localhost>]

On Thu, 4 Sep 2008, Robert Elz wrote:

>    Date:        Thu, 4 Sep 2008 05:47:27 -0700 (PDT)
>    From:        Paul Goyette <paul%whooppee.com@localhost>
>    Message-ID:  
> <Pine.NEB.4.64.0809040545050.23592%quicky.whooppee.com@localhost>
>
>  | Hmmm, during sysinst?  I'll have to check into that.  I wouldn't expect
>  | to see them during sysinst;  I would only expect them on systems that
>  | have actual monitor values being set in the /etc/envsys.cfg file.  If
>  | the messages are showing up other cases, that would be a bug.  :)
>
> I'm not getting impi messages, but since I upgraded to a recent -current
> (about 5 days old now) I've started seeing what I presume is the ACPI
> equivalent ...
>
>         coretemp0: normal state on 'cpu0 temperature'
>         acpitz0: normal state on 'temperature'
>         acpitz0: normal state on 'temperature'
>         coretemp1: normal state on 'cpu1 temperature'
>
> I don't have an /etc/envsys.cfg file, I do seem to have /etc/envsys.conf
> but it contains nothing aside from what I assume are comments (# lines).
> I certainly have never configured it.

Yeah, that was a typo on my part.  I meant '.conf'.

Since your file contains only comments, then I will take a look and see 
why the messages are getting reported.

> I'd prefer if this kind of noise never appeared on the console at all,
> unless it is reporting some kind of emergency situation, if you need to
> send it to syslog, I guess that's OK, but the console should really
> never report anything that isn't indicating something like "cpu on fire"
> or "disc drive 3 just fell out of the tray" - at least, not after the
> initial boot attach messages, and even those tend to say way too much
> (just telling me the hardware has a driver attached (or not) is all I
> need to know, everything else should be available some other way).

The actual console messages are created by the scripts in 
/etc/powerd/scripts directory.  For normal (ie, non-INSTALL) situations 
you can easily modify them.

I will certainly investigate making the "verbosity" of sysmon_envsys(9) 
more tailorable as part of a revamp that I'm planning for the post-5.0 
timeframe.

> I've also noticed the recent kernel now instists on telling me ...
>
> WARNING: the union file system is experimental and may be unstable
> WARNING: the union file system is experimental and may be unstable
>
> which comes from (I am assuming)
>
> /usr/pkg/lib/X11/app-defaults /usr/X11R6/lib/X11/app-defaults union rw,hidden
> /var/xkb /usr/X11R6/lib/X11/xkb union rw,hidden
>
> in /etc/fstab.   I know union sometimes has problems.   I also know that
> these 2 mounts never hit any of them.   I don't need the kernel bitching
> at me.   If it is necessary to warn somewhere, have mount_union do it,
> not the kernel - that way, I can just direct output at /dev/null and
> ignore it...

HeHeHe!  Unionfs is (fortunately) someone else's problem!  :)


----------------------------------------------------------------------
|   Paul Goyette   | PGP DSS Key fingerprint: |  E-mail addresses:   |
| Customer Service | FA29 0E3B 35AF E8AE 6651 |  paul%whooppee.com@localhost   |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette%juniper.net@localhost |
----------------------------------------------------------------------