On Linux, verify that the kernel entropy pool is capable of providing
strong random data before allowing WPA/WPA2 connection to be
established. If 20 bytes of data cannot be read from /dev/random,
force first two 4-way handshakes to fail while collecting entropy
into the internal pool in hostapd. After that, give up on /dev/random
and allow the AP to function based on the combination of /dev/urandom
and whatever data has been collected into the internal entropy pool.
wlan0: RADIUS No authentication server configured
MEMLEAK[0x999feb8]: len 1040
WPA_TRACE: memleak - START
[3]: ./hostapd(radius_msg_new+0x33) [0x8074f43]
radius_msg_new() ../src/radius/radius.c:117
[4]: ./hostapd() [0x806095e]
ieee802_1x_encapsulate_radius() ../src/ap/ieee802_1x.c:439
ieee802_1x_aaa_send() ../src/ap/ieee802_1x.c:1496
For example, this error occured when I used WPS hostapd without
"eap_server=1" definition in configuration file.
By default, make hostapd and wpa_supplicant maintain an internal
entropy pool that is fed with following information:
hostapd:
- Probe Request frames (timing, RSSI)
- Association events (timing)
- SNonce from Supplicants
wpa_supplicant:
- Scan results (timing, signal/noise)
- Association events (timing)
The internal pool is used to augment the random numbers generated
with the OS mechanism (os_get_random()). While the internal
implementation is not expected to be very strong due to limited
amount of generic (non-platform specific) information to feed the
pool, this may strengthen key derivation on some devices that are
not configured to provide strong random numbers through
os_get_random() (e.g., /dev/urandom on Linux/BSD).
This new mechanism is not supposed to replace proper OS provided
random number generation mechanism. The OS mechanism needs to be
initialized properly (e.g., hw random number generator,
maintaining entropy pool over reboots, etc.) for any of the
security assumptions to hold.
If the os_get_random() is known to provide strong ramdom data (e.g., on
Linux/BSD, the board in question is known to have reliable source of
random data from /dev/urandom), the internal hostapd random pool can be
disabled. This will save some in binary size and CPU use. However, this
should only be considered for builds that are known to be used on
devices that meet the requirements described above. The internal pool
is disabled by adding CONFIG_NO_RANDOM_POOL=y to the .config file.
This commit adds a new wrapper, random_get_bytes(), that is currently
defined to use os_get_random() as is. The places using
random_get_bytes() depend on the returned value being strong random
number, i.e., something that is infeasible for external device to
figure out. These values are used either directly as a key or as
nonces/challenges that are used as input for key derivation or
authentication.
The remaining direct uses of os_get_random() do not need as strong
random numbers to function correctly.
This adds more time for the system entropy pool to be filled before
requesting random data for generating the WPA/WPA2 encryption keys.
This can be helpful especially on embedded devices that do not have
hardware random number generator and may lack good sources of
randomness especially early in the bootup sequence when hostapd is
likely to be started.
GMK and Key Counter are still initialized once in the beginning to
match the RSN Authenticator state machine behavior and to make sure
that the driver does not transmit broadcast frames unencrypted.
However, both GMK (and GTK derived from it) and Key Counter will be
re-initialized when the first station connects and is about to
enter 4-way handshake.
The example GMK-to-GTK derivation described in the IEEE 802.11 standard
is marked informative and there is no protocol reason for following it
since this derivation is done only on the AP/Authenticator and does not
need to match with the Supplicant. Mix in more data into the derivation
process to get more separation from GMK.
The previous commit broke completion in various places. The proper
way of handling the completion of full word is to verify whether
there are more than one possible match at that point.
Completion needs to be done even if the full word has been entered.
In addition, fix the space-after-full-word to properly allocate room
for the extra character when completion is used in the middle of the
string.
This adds new commands for wlantest_cli to request wlantest to
inject frames. This version can only send out Authentication
frames and unprotected SA Query Request frames, but there is
now place to add more frames and encryption with future commits.
ap_setup_locked=2 can now be used to enable a special mode where
WPS ER can learn the current AP settings, but cannot change then.
In other words, the protocol is allowed to continue past M2, but
is stopped at M7 when AP is in this mode. WPS IE does not
advertise AP Setup Locked in this case to avoid interoperability
issues.
In wpa_supplicant, use ap_setup_locked=2 by default. Since the AP PIN
is disabled by default, this does not enable any new functionality
automatically. To allow the read-only ER to go through the protocol,
wps_ap_pin command needs to be used to enable the AP PIN.
There is no point in maintaining two almost identical versions
of this parser. Move WPA IE parser into wpa_common.c similarly
to what was already the case with RSN IE parse.
When multiple wireless interfaces are used with WPS, the UPnP
subscriptions need to be removed whenever a matching Registrar
instance gets removed. This avoids a segfault due to access to
freed memory during hostapd shutdown.
In addition, the UPnP interface instance structure needs to be
freed to avoid memory leak.
WPS external Registrars can get confused about multiple UPnP
instances (one per radio) on a dual-concurrent APs. Simplify the
design by sharing a single UPnP state machine for all wireless
interfaces controlled by hostapd. This matches with the previous
changes that made a single command enable WPS functionality on
all interfaces.
This is relatively minimal change to address the sharing of the
state among multiple struct hostapd_data instances. More cleanup
can be done separately to remove unnecessary copies of information.
When the peer does not include our initial preference in the Channel
List attribute during GO Negotiation, try to use the best channel of
the other band as the new preference instead of falling back to the
first channel in the intersection.
The driver wrapper may now indicate the preferred channel (e.g., based
on scan results) on both 2.4 GHz and 5 GHz bands (and an overall best
frequency). When setting up a GO, this preference information is used
to select the operating channel if configuration does not include
hardcoded channel. Similarly, this information can be used during
GO Negotiation to indicate preference for a specific channel based
on current channel conditions.
p2p_group_add command can now use special values (freq=2 and freq=5)
to indicate that the GO is to be started on the specified band.
Commit d8d940b746 introduced a regression
that prevented TSN APs from being used with WEP since the AP was
rejected if it advertised WPA or RSN IE when we were configured to use
WEP. Resolve this by checking whether the AP is advertising a TSN, i.e.,
whether the AP allows WEP to be used as a group cipher. If so, allow
the AP to be selected if we are configured to use static WEP or
IEEE 802.1X (non-WPA).
It should be noted that this is still somewhat more restricted in AP
selection than earlier wpa_supplicant branches (0.7.x or older) that
ignore the WPA/RSN IE completely when configured for non-WPA.
