On Wed, Jan 7, 2015 at 12:11 PM, William Hubbs <willi...@gentoo.org> wrote: > On Wed, Jan 07, 2015 at 11:21:56AM -0500, Mike Pagano wrote: >> On Tue, Jan 06, 2015 at 05:47:10PM -0600, William Hubbs wrote: >> > All, >> > >> > these packages have been masked in the tree for months - years with no >> > signs of fixes. >> > >> > I am particularly concerned about packages with known security >> > vulnerabilities staying in the main tree masked. If people want to keep >> > using those packages, I don't want to stop them, but packages like this >> > should not be in the main tree. >> > >> > # Mask gentoo-sources ebuilds that are affected with security bug >> > CVE-2014-3153. >> > # >> > # Pinkie Pie discovered an issue in the futex subsystem that allows a >> > # local user to gain ring 0 control via the futex syscall. An >> > # unprivileged user could use this flaw to crash the kernel (resulting >> > # in denial of service) or for privilege escalation. >> > # >> > # https://bugs.gentoo.org/show_bug.cgi?id=CVE-2014-3153 >> > =sys-kernel/gentoo-sources-3.2.58-r2 >> > ~sys-kernel/gentoo-sources-3.4.90 >> > =sys-kernel/gentoo-sources-3.4.91 >> > ~sys-kernel/gentoo-sources-3.10.40 >> > =sys-kernel/gentoo-sources-3.10.41 >> > ~sys-kernel/gentoo-sources-3.12.20 >> > =sys-kernel/gentoo-sources-3.12.21 >> > ~sys-kernel/gentoo-sources-3.14.4 >> > =sys-kernel/gentoo-sources-3.14.5 > > Mike, > > since you responded here, what do you think about this p.mask entry? > Should we keep these in the tree? > >> >> Hello, >> >> What's the feeling for how long a package.mask entry should stay in the >> file in the event that a package can cause physical damage to a user's >> system. >> >> For certain types of hardware, kernel 3.17.0 could cause some >> filesystem corruption. Of couse, 3.17.0 is out of the tree but when is >> it appropiate to say that a user has had enough time to upgarde their >> systems and we can remove this entry? > > (qa hat off here, just a question) > > I'm a bit confused here. > If you have a specific p.mask entry for 3.17.0 and 3.17.0 is out of the > tree, isn't that p.mask entry invalid now? If so, go ahead and remove > or adjust the entry. >
If users currently have 3.17.0 installed, portage will output a warning message about a masked package being installed, even if the ebuild no longer exists in the tree. If you remove the mask, users will no longer be warned that they are using a flawed copy of the kernel sources. Thus, Mike's question about timing.
