On Wed, 2019-06-05 at 17:14 -0400, Michael Orlitzky wrote:
> On 6/5/19 5:12 AM, Michał Górny wrote:
> > +
> > + # check for ACCT_USER_ID collisions early
> > + if [[ -n ${ACCT_USER_ENFORCE_ID} ]]; then
> > + local pwd=$(egetent passwd "${ACCT_USER_ID}")
> > + if [[ -n ${pwd} ]]; then
> > + eerror "The required UID is already taken by another
> > user."
> > + eerror " UID: ${ACCT_USER_ID} (needed for ${PN})"
> > + eerror " current user: ${pwd}"
> > + die "UID ${ACCT_USER_ID} taken already"
> > + fi
> > + fi
> > +}
>
> If we set ACCT_USER_ENFORCE_ID=true in ftp-0.ebuild and then "emerge
> acct-user/ftp" again, the second one dies with
>
> * The required UID is already taken by another user.
> * UID: 21 (needed for ftp)
>
> This prevents "emerge -e @world" from working, and would prevent an
> upgrade to ftp-1.ebuild in the future.
>
> We could augment the checks to ignore the existing user/group if its
> name agrees with the one we're trying to create, but now I'm having deja
> vu. I think this is another reason why I decided to go with a dummy file
> installed under /var/lib to catch collisions: the PM will necessarily
> ignore collisions from the same package.Good catch. Let's see what others have to say. If we're going to check for username patches, we'd need another user.eclass portability knob. -- Best regards, Michał Górny
signature.asc
Description: This is a digitally signed message part
