[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: deluser on purge (was: Piuparts testing status update)



On Tue, 14 Nov 2006, Russ Allbery wrote:
> This is something that I'd really like to see us sort out in policy,
> since I think we should be able to describe consistent behavior with
> regard to system users and package purging to our users.

What makes the most sense to me is to not delete the user, and warn
that this has not been done. (I'm really not sure how best to do the
warning besides outputing to STDERR.)

This avoids the obvious problems with deleting a user who may still
own files on the system, and then recreating a different username for
a different program with the same uid which shouldn't have access to
those files (or, worse, if someone was insane and made something
setuid to the autogenerated uid.)

A further refinement of this suggestion is to allow/suggest prompting
using debconf with a low priority question to remove the user, with
the default set to not delete. [This would be my personal preference;
it may even be worthwhile to consider codifying a best practice,[1]
and then if Joey Hess agrees, creating a dh_installuser or similar
script which implements it, including debconf routines.]

This would allow individuals who knew that they wanted to delete the
user to easily cause the user to be deleted, and do so in an automated
fashion.


Don Armstrong

1: Granted, this best practice should probaly be codified in the
Developer's Reference, not policy, but we could discuss it at the same
time.
-- 
It was said that life was cheap in Ankh-Morpork. This was, of course,
completely wrong. Life was often very expensive; you could get death
for free.
 -- Terry Pratchet _Pyramids_ p25

http://www.donarmstrong.com              http://rzlab.ucr.edu



Reply to: