On Mon, Nov 26, 2001 at 12:09:49PM +1100, Daniel Stone wrote:

> > > That looks like acceptable behaviour to me; the script was looking for a
> > > file that wasn't there and it told you exactly that.  When you put the
> > > file back (or a dummy one at least) it worked OK.  What exactly about the
> > > process do you not think is robust?
> > 
> > The file was in the target state (deleted). The script (and thus the package
> > management system, because it will put this package in a broken state)
> > should not b0rk because it is unable to delete a file that is "already not
> > there".
> 
> It was in the prerm script, thus the package wasn't removed yet. prerm
> == pre-removal, so nothing had yet been done. If the prerm script isn't
> run, then purge/whatever may do completely unexpected things, so that's
> a completely legitimate thing to do. The script needs to run in prerm,
> before removal (oddly enough). You're applying for NM, you should know
> this. Read up on policy.

Just so I make sure I'm following, this makes that the prerm
script is broken, and not apt, right?  (prerm should have ignored
the exit code from rm.)


-- 
#ozone/algorithm <[EMAIL PROTECTED]>          - trust.in.love.to.save

-- 
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Reply via email to