Skip to content

RFC access checks for translating contexts

Darrel Goeddel selinux 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest3 patches 10 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/3reference policy: add "context" security classpolicy/flask/access_vectors policy/flask/security_classes policy/mls +1 more+0 -07 replies
2/3libselinux: define the "context" security classlibselinux/include/selinux/av_permissions.h libselinux/include/selinux/flask.h libselinux/src/av_perm_to_string.h +1 more+3 -02 replies
3/3mcstransd: perform an access check on the conext to be translatedsrc/mcstransd.c+0 -01 reply

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [RFC PATCH 0/3] access checks for translating contexts

The following is an attempt to perform access checks for context translations. The idea being that a process should not know about labels that are outside of it's clearance. Since there are now standalone MLS checks available, I have added a new security class "context" with permission "translate". The mlsconstraint on that permission handles the MLS clearance portion. TE access must also be granted for the context to be translated - I see this a drawback of the implementation because now we need a way to give TE access to all types if we want a process to do translations limited purely by MLS.

Now... The daemon running at the lowest MLS level and the file describing translations is at the lowest MLS level. This throws the whole idea of protecting the labels (the reason for the daemon in the first place) themselves out the door since everyone can just read the file. That daemon needs to run at the highest MLS level and the file needs to be at the highest MLS level. We (TCS) had things set up that way when we did some of the initial work on the daemon. Has anyone looked into actually fixing this issue (or at least have an idea on what caused the breakage)? If not, this whole patchset is really not necessary. I guess I could look into that as well...

Signature · 6 lines
Darrel

--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to [email protected] with
the words "unsubscribe selinux" without quotes as the message.