1 Device Whitelist Controller
5 Implement a cgroup to track and enforce open and mknod restrictions
6 on device files. A device cgroup associates a device access
7 whitelist with each cgroup. A whitelist entry has 4 fields.
8 'type' is a (all), c (char), or b (block). 'all' means it applies
9 to all types and all major and minor numbers. Major and minor are
10 either an integer or * for all. Access is a composition of r
11 (read), w (write), and m (mknod).
13 The root device cgroup starts with rwm to 'all'. A child device
14 cgroup gets a copy of the parent. Administrators can then remove
15 devices from the whitelist or add new entries. A child cgroup can
16 never receive a device access which is denied by its parent.
20 An entry is added using devices.allow, and removed using
21 devices.deny. For instance
23 echo 'c 1:3 mr' > /sys/fs/cgroup/1/devices.allow
25 allows cgroup 1 to read and mknod the device usually known as
28 echo a > /sys/fs/cgroup/1/devices.deny
30 will remove the default 'a *:* rwm' entry. Doing
32 echo a > /sys/fs/cgroup/1/devices.allow
34 will add the 'a *:* rwm' entry to the whitelist.
38 Any task can move itself between cgroups. This clearly won't
39 suffice, but we can decide the best way to adequately restrict
40 movement as people get some experience with this. We may just want
41 to require CAP_SYS_ADMIN, which at least is a separate bit from
42 CAP_MKNOD. We may want to just refuse moving to a cgroup which
43 isn't a descendant of the current one. Or we may want to use
44 CAP_MAC_ADMIN, since we really are trying to lock down root.
46 CAP_SYS_ADMIN is needed to modify the whitelist or move another
47 task to a new cgroup. (Again we'll probably want to change that).
49 A cgroup may not be granted more permissions than the cgroup's
54 device cgroups maintain hierarchy by making sure a cgroup never has more
55 access permissions than its parent. Every time an entry is written to
56 a cgroup's devices.deny file, all its children will have that entry removed
57 from their whitelist and all the locally set whitelist entries will be
58 re-evaluated. In case one of the locally set whitelist entries would provide
59 more access than the cgroup's parent, it'll be removed from the whitelist.
66 group behavior exceptions
67 A allow "b 8:* rwm", "c 116:1 rw"
68 B deny "c 1:3 rwm", "c 116:2 rwm", "b 3:* rwm"
70 If a device is denied in group A:
71 # echo "c 116:* r" > A/devices.deny
72 it'll propagate down and after revalidating B's entries, the whitelist entry
73 "c 116:2 rwm" will be removed:
75 group whitelist entries denied devices
76 A all "b 8:* rwm", "c 116:* rw"
77 B "c 1:3 rwm", "b 3:* rwm" all the rest
79 In case parent's exceptions change and local exceptions are not allowed
80 anymore, they'll be deleted.
82 Notice that new whitelist entries will not be propagated:
87 group whitelist entries denied devices
88 A "c 1:3 rwm", "c 1:5 r" all the rest
89 B "c 1:3 rwm", "c 1:5 r" all the rest
91 when adding "c *:3 rwm":
92 # echo "c *:3 rwm" >A/devices.allow
95 group whitelist entries denied devices
96 A "c *:3 rwm", "c 1:5 r" all the rest
97 B "c 1:3 rwm", "c 1:5 r" all the rest
99 but now it'll be possible to add new entries to B:
100 # echo "c 2:3 rwm" >B/devices.allow
101 # echo "c 50:3 r" >B/devices.allow
103 # echo "c *:3 rwm" >B/devices.allow
105 Allowing or denying all by writing 'a' to devices.allow or devices.deny will
106 not be possible once the device cgroups has children.
108 4.1 Hierarchy (internal implementation)
110 device cgroups is implemented internally using a behavior (ALLOW, DENY) and a
111 list of exceptions. The internal state is controlled using the same user
112 interface to preserve compatibility with the previous whitelist-only
113 implementation. Removal or addition of exceptions that will reduce the access
114 to devices will be propagated down the hierarchy.
115 For every propagated exception, the effective rules will be re-evaluated based
116 on current parent's access rules.