Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-24 Thread Anselm R Garbe
On Tue, Jan 24, 2012 at 11:01:19AM +0100, Michael Stummvoll wrote: > Hi, > > I would be willing to apply the slock.c patch, if you can supply diff > > -u or hg diff output accordingly. > > thanks, patch is attached. Oh I see, it is against the 0.9 release and not against tip, nevermind... Cheers

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-24 Thread Michael Stummvoll
Hi, > I would be willing to apply the slock.c patch, if you can supply diff > -u or hg diff output accordingly. thanks, patch is attached. Kind Regards, Michael --- slock.c.orig 2012-01-24 10:57:52.751270668 +0100 +++ slock.c 2012-01-24 10:59:00.616019472 +0100 @@ -38,9 +38,9 @@ const char *rva

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Anselm R Garbe
Hi Michael, On 23 January 2012 15:49, Michael Stummvoll wrote: > On 23.01.2012 15:40, Connor Lane Smith wrote: >> On 23 January 2012 14:38, Nick wrote: >>> No no no, much simpler that debian holds that 1 line >>> Makefile patch, and any other distributions who want to >>> change it from setuid.

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Michael Stummvoll
On 23.01.2012 15:40, Connor Lane Smith wrote: > Hey, > > On 23 January 2012 14:38, Nick wrote: >> No no no, much simpler that debian holds that 1 line >> Makefile patch, and any other distributions who want to >> change it from setuid. Such things are very distro-specific, >> and besides, changing

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Connor Lane Smith
Hey, On 23 January 2012 14:38, Nick wrote: > No no no, much simpler that debian holds that 1 line > Makefile patch, and any other distributions who want to > change it from setuid. Such things are very distro-specific, > and besides, changing a config.mk setting in a debian patch > is EXACTLY as

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Nick
On Mon, Jan 23, 2012 at 03:35:02PM +0100, Michael Stummvoll wrote: > than, may we could make the shadow-group configurable in the config.mk or > Makefile? No no no, much simpler that debian holds that 1 line Makefile patch, and any other distributions who want to change it from setuid. Such thing

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Michael Stummvoll
than, may we could make the shadow-group configurable in the config.mk or Makefile? Kind Regards, Michael

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Kurt H Maier
On Mon, Jan 23, 2012 at 03:20:44PM +0100, Michael Stummvoll wrote: > I don't think, that checking if the tool has access to the password instead > of assuming this by its effective user is distribution specific. It isn't. Using setgid shadow is.

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Michael Stummvoll
Hi, I don't think, that checking if the tool has access to the password instead of assuming this by its effective user is distribution specific. Kind Regards, Michael

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Kurt H Maier
On Mon, Jan 23, 2012 at 03:03:07PM +0100, Michael Stummvoll wrote: > With the Makefile your object may is right. Maybe somebody finds a better > solution here than mine The better solution is for distribution packagers to stop trying to push distribution-specific garbage upstream. If there's a "

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Michael Stummvoll
On 23.01.2012 14:28, Kurt H Maier wrote: > Not all distributions even have a shadow group. the patch in slock.c is indepedent of this group, it just checks if it can get the password-entry, so it still runs on systems without shadow-group. It doesn't matter at all if you realize the access with

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Stephen Paul Weber
Somebody claiming to be Kurt H Maier wrote: On Mon, Jan 23, 2012 at 01:33:35PM +0100, Michael Stummvoll wrote: so I patched slock to not demant root-rights, but just checks, if the password-query commands are successfull. I also patched the Makefile to do not setuid root but setgid shadow instea

Re: [dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Kurt H Maier
On Mon, Jan 23, 2012 at 01:33:35PM +0100, Michael Stummvoll wrote: > I fresh adopted the maintaince for the suckless-tools package in debian. > Its debian-policy to avoid setuid root binaries wherever this is possible, > so I patched slock to not demant root-rights, but just checks, > if the passwo

[dev] [slock] patch for avoiding hardcoded root-check

2012-01-23 Thread Michael Stummvoll
Hi suckless folks, I fresh adopted the maintaince for the suckless-tools package in debian. Its debian-policy to avoid setuid root binaries wherever this is possible, so I patched slock to not demant root-rights, but just checks, if the password-query commands are successfull. I also patched the M