xref: /linux/Documentation/security/keys/request-key.rst (revision ae22a94997b8a03dcb3c922857c203246711f9d4)
1===================
2Key Request Service
3===================
4
5The key request service is part of the key retention service (refer to
6Documentation/security/keys/core.rst).  This document explains more fully how
7the requesting algorithm works.
8
9The process starts by either the kernel requesting a service by calling
10``request_key*()``::
11
12	struct key *request_key(const struct key_type *type,
13				const char *description,
14				const char *callout_info);
15
16or::
17
18	struct key *request_key_tag(const struct key_type *type,
19				    const char *description,
20				    const struct key_tag *domain_tag,
21				    const char *callout_info);
22
23or::
24
25	struct key *request_key_with_auxdata(const struct key_type *type,
26					     const char *description,
27					     const struct key_tag *domain_tag,
28					     const char *callout_info,
29					     size_t callout_len,
30					     void *aux);
31
32or::
33
34	struct key *request_key_rcu(const struct key_type *type,
35				    const char *description,
36				    const struct key_tag *domain_tag);
37
38Or by userspace invoking the request_key system call::
39
40	key_serial_t request_key(const char *type,
41				 const char *description,
42				 const char *callout_info,
43				 key_serial_t dest_keyring);
44
45The main difference between the access points is that the in-kernel interface
46does not need to link the key to a keyring to prevent it from being immediately
47destroyed.  The kernel interface returns a pointer directly to the key, and
48it's up to the caller to destroy the key.
49
50The request_key_tag() call is like the in-kernel request_key(), except that it
51also takes a domain tag that allows keys to be separated by namespace and
52killed off as a group.
53
54The request_key_with_auxdata() calls is like the request_key_tag() call, except
55that they permit auxiliary data to be passed to the upcaller (the default is
56NULL).  This is only useful for those key types that define their own upcall
57mechanism rather than using /sbin/request-key.
58
59The request_key_rcu() call is like the request_key_tag() call, except that it
60doesn't check for keys that are under construction and doesn't attempt to
61construct missing keys.
62
63The userspace interface links the key to a keyring associated with the process
64to prevent the key from going away, and returns the serial number of the key to
65the caller.
66
67
68The following example assumes that the key types involved don't define their
69own upcall mechanisms.  If they do, then those should be substituted for the
70forking and execution of /sbin/request-key.
71
72
73The Process
74===========
75
76A request proceeds in the following manner:
77
78  1) Process A calls request_key() [the userspace syscall calls the kernel
79     interface].
80
81  2) request_key() searches the process's subscribed keyrings to see if there's
82     a suitable key there.  If there is, it returns the key.  If there isn't,
83     and callout_info is not set, an error is returned.  Otherwise the process
84     proceeds to the next step.
85
86  3) request_key() sees that A doesn't have the desired key yet, so it creates
87     two things:
88
89      a) An uninstantiated key U of requested type and description.
90
91      b) An authorisation key V that refers to key U and notes that process A
92     	 is the context in which key U should be instantiated and secured, and
93     	 from which associated key requests may be satisfied.
94
95  4) request_key() then forks and executes /sbin/request-key with a new session
96     keyring that contains a link to auth key V.
97
98  5) /sbin/request-key assumes the authority associated with key U.
99
100  6) /sbin/request-key execs an appropriate program to perform the actual
101     instantiation.
102
103  7) The program may want to access another key from A's context (say a
104     Kerberos TGT key).  It just requests the appropriate key, and the keyring
105     search notes that the session keyring has auth key V in its bottom level.
106
107     This will permit it to then search the keyrings of process A with the
108     UID, GID, groups and security info of process A as if it was process A,
109     and come up with key W.
110
111  8) The program then does what it must to get the data with which to
112     instantiate key U, using key W as a reference (perhaps it contacts a
113     Kerberos server using the TGT) and then instantiates key U.
114
115  9) Upon instantiating key U, auth key V is automatically revoked so that it
116     may not be used again.
117
118  10) The program then exits 0 and request_key() deletes key V and returns key
119      U to the caller.
120
121This also extends further.  If key W (step 7 above) didn't exist, key W would
122be created uninstantiated, another auth key (X) would be created (as per step
1233) and another copy of /sbin/request-key spawned (as per step 4); but the
124context specified by auth key X will still be process A, as it was in auth key
125V.
126
127This is because process A's keyrings can't simply be attached to
128/sbin/request-key at the appropriate places because (a) execve will discard two
129of them, and (b) it requires the same UID/GID/Groups all the way through.
130
131
132Negative Instantiation And Rejection
133====================================
134
135Rather than instantiating a key, it is possible for the possessor of an
136authorisation key to negatively instantiate a key that's under construction.
137This is a short duration placeholder that causes any attempt at re-requesting
138the key while it exists to fail with error ENOKEY if negated or the specified
139error if rejected.
140
141This is provided to prevent excessive repeated spawning of /sbin/request-key
142processes for a key that will never be obtainable.
143
144Should the /sbin/request-key process exit anything other than 0 or die on a
145signal, the key under construction will be automatically negatively
146instantiated for a short amount of time.
147
148
149The Search Algorithm
150====================
151
152A search of any particular keyring proceeds in the following fashion:
153
154  1) When the key management code searches for a key (keyring_search_rcu) it
155     firstly calls key_permission(SEARCH) on the keyring it's starting with,
156     if this denies permission, it doesn't search further.
157
158  2) It considers all the non-keyring keys within that keyring and, if any key
159     matches the criteria specified, calls key_permission(SEARCH) on it to see
160     if the key is allowed to be found.  If it is, that key is returned; if
161     not, the search continues, and the error code is retained if of higher
162     priority than the one currently set.
163
164  3) It then considers all the keyring-type keys in the keyring it's currently
165     searching.  It calls key_permission(SEARCH) on each keyring, and if this
166     grants permission, it recurses, executing steps (2) and (3) on that
167     keyring.
168
169The process stops immediately a valid key is found with permission granted to
170use it.  Any error from a previous match attempt is discarded and the key is
171returned.
172
173When request_key() is invoked, if CONFIG_KEYS_REQUEST_CACHE=y, a per-task
174one-key cache is first checked for a match.
175
176When search_process_keyrings() is invoked, it performs the following searches
177until one succeeds:
178
179  1) If extant, the process's thread keyring is searched.
180
181  2) If extant, the process's process keyring is searched.
182
183  3) The process's session keyring is searched.
184
185  4) If the process has assumed the authority associated with a request_key()
186     authorisation key then:
187
188      a) If extant, the calling process's thread keyring is searched.
189
190      b) If extant, the calling process's process keyring is searched.
191
192      c) The calling process's session keyring is searched.
193
194The moment one succeeds, all pending errors are discarded and the found key is
195returned.  If CONFIG_KEYS_REQUEST_CACHE=y, then that key is placed in the
196per-task cache, displacing the previous key.  The cache is cleared on exit or
197just prior to resumption of userspace.
198
199Only if all these fail does the whole thing fail with the highest priority
200error.  Note that several errors may have come from LSM.
201
202The error priority is::
203
204	EKEYREVOKED > EKEYEXPIRED > ENOKEY
205
206EACCES/EPERM are only returned on a direct search of a specific keyring where
207the basal keyring does not grant Search permission.
208