Subject: RE: FS behaviour on error
To: None <"''">
From: Andy Sporner <>
List: tech-kern
Date: 04/10/2000 15:10:56
Since I am new to this group I have a question with respect
to this.

I see many suggestions about changes to BSD (I am not sure how
much of a change this is--if any).

How sacred is the implementation of BSD?  I have posted a
request for comments about PID's and heard nothing so I am
presuming that there isn't a big religous issue about possible

What exactly is the stand of the BSD group?  I happen to 
think this idea below is a good thing if we don't have it,
with respect to filesystem behaviour.

I apologize for some of any of my possible ignorance.  I have
been mostly in the Linux arena, and previously mostly System/V,
so I am still working on the particulars of BSD, so please be



> -----Original Message-----
> From: Hubert Feyrer []
> Sent: Monday, April 10, 2000 11:04 AM
> To:
> Subject: FS behaviour on error
> It's been discussed previously what the correct behaviour is when a
> filesystem error is discovered. To provide some input, here's what
> mount)ufs on Solaris 7 has to say about this:
>                onerror=action
>                          This option specifies  the  action  that
>                          UFS  should  take  to  recover  from  an
>                          internal inconsistency on a file system.
>                          Specify action as panic, lock, or umount
>                          . These values  cause  a  forced  system
>                          shutdown,  a  file  system  lock  to  be
>                          applied to the file system, or the  file
>                          system to be forcibly unmounted, respec-
>                          tively. The default is panic.
>  - Hubert
> -- 
> Microsoft: "Where do you want to go today?"
> Linux:     "Where do you want to be tomorrow?"
> BSD:       "Are you guys coming, or what?"