Package: debian-policy
Version: 3.5.2.0
Severity: wishlist
The man page for dh_suidregister says that any package containing a
SUID/SGID binary no longer needs to use suidregister, instead, users
can use dpkg-statoverride as necessary.
Policy (section 11.9, "Permissions and Owners") doesn't talk
On Sun, Mar 11, 2001 at 07:03:51PM +0100, Josip Rodin wrote:
> > > > I'm packaging viewmol, a program for visualising molecules, used in
> > > computational chemistry.
> > >
> > > The question arises, what subsection should it be put into?
> > > A comparable program, rasmol, is already in debian,
On Tue, Mar 13, 2001 at 10:54:45PM +1100, Drew Parsons wrote:
> Seems to me there's still a gaping hole in policy though. Is
> packages.debian.org really supposed to be the definitive arbiter of policy
> with regards to sections? The grep-available trick is obscure enough to not
> count. Shouldn
On Mon, Mar 12, 2001 at 10:20:49PM -0800, Ben Gertzfield wrote:
> Package: debian-policy
> Version: 3.5.2.0
> Severity: wishlist
>
> The man page for dh_suidregister says that any package containing a
> SUID/SGID binary no longer needs to use suidregister, instead, users
> can use dpkg-statoverrid
Previously Ben Gertzfield wrote:
> If I understand dpkg-statoverride correctly, mentioning that
> dpkg-statoverride is not to be called from maintainer post/pre
> install scripts should also be added.
Euhm, how did you come to that conclusion?
Wichert.
--
On Mon, 12 Mar 2001, Ben Gertzfield wrote:
> If I understand dpkg-statoverride correctly, mentioning that
> dpkg-statoverride is not to be called from maintainer post/pre
> install scripts should also be added.
I oppose that. How am I supposed to have dynamically-created UIDs [that are
not created
On Tue, Mar 13, 2001 at 11:56:59AM -0300, Henrique M Holschuh wrote:
> On Mon, 12 Mar 2001, Ben Gertzfield wrote:
> > If I understand dpkg-statoverride correctly, mentioning that
> > dpkg-statoverride is not to be called from maintainer post/pre
> > install scripts should also be added.
>
> I oppo
On Tue, Mar 13, 2001 at 02:21:47PM +0100, Wichert Akkerman wrote:
> Previously Ben Gertzfield wrote:
> > If I understand dpkg-statoverride correctly, mentioning that
> > dpkg-statoverride is not to be called from maintainer post/pre
> > install scripts should also be added.
>
> Euhm, how did you c
Previously Julian Gilbey wrote:
> dpkg-statoverride --list is OK, adding or removing overrides is almost
> certainly not.
It most certainly is. Think dynamic useres and groups, user interaction
if debconf isn't available, etc.
Wichert.
--
On Tue, 13 Mar 2001, Julian Gilbey wrote:
> On Tue, Mar 13, 2001 at 11:56:59AM -0300, Henrique M Holschuh wrote:
> > On Mon, 12 Mar 2001, Ben Gertzfield wrote:
> > > If I understand dpkg-statoverride correctly, mentioning that
> > > dpkg-statoverride is not to be called from maintainer post/pre
> >
Previously Julian Gilbey wrote:
> But you mustn't use dpkg-statoverride to introduce new overrides;
> that's the responsibility of the local sysadmin.
That is a vlid point, but has nothing to do with where you use
dpkg-statoverride.
I don't think policy should mention where you can use it. Policy
> "Wichert" == Wichert Akkerman <[EMAIL PROTECTED]> writes:
Julian> dpkg-statoverride --list is OK, adding or removing
Julian> overrides is almost certainly not.
Wichert> It most certainly is. Think dynamic useres and groups,
Wichert> user interaction if debconf isn't availabl
> "Wichert" == Wichert Akkerman <[EMAIL PROTECTED]> writes:
Wichert> Policy should set guidelines for making packages [...]
Wichert> The less details, the better.
Um. Policy *IS* the guide for making packages now. There's no
Packaging Manual any more, and so these kinds of details
On Tue, Mar 13, 2001 at 10:54:45PM +1100, Drew Parsons wrote:
> Seems to me there's still a gaping hole in policy though. Is
> packages.debian.org really supposed to be the definitive arbiter of policy
> with regards to sections? The grep-available trick is obscure enough to not
> count. Shouldn
Previously Ben Gertzfield wrote:
> That's a good usage, but for the "old" suidregister-like behavior,
> normal SUID/SGID apps do *not* need to be dpkg-statoverride'd,
> as far as I can tell. That's all I want to get in policy; we
> should mention that it's OK for dynamic users/groups.
What exactl
On 20010313T203223+0100, Josip Rodin wrote:
> Well, none of these are the canonical source, they just read data from it --
> the override file, see the /indices/override.* files on every mirror.
Actually, the canonical source is Katie's database. Even the override
files are nowadays generated fro
On Tue, Mar 13, 2001 at 08:32:23PM +0100, Josip Rodin wrote:
> On Tue, Mar 13, 2001 at 10:54:45PM +1100, Drew Parsons wrote:
> > Seems to me there's still a gaping hole in policy though. Is
> > packages.debian.org really supposed to be the definitive arbiter of policy
> > with regards to sections?
> "Wichert" == Wichert Akkerman <[EMAIL PROTECTED]> writes:
Wichert> What exactly do you want in policy? `if you behave like
Wichert> the old suidregister don't do this' or so? I really don't
Wichert> see the use of that.
The conflicts that man dh_suidregister mentions should be i
18 matches
Mail list logo
