This uses similar, but not identical, interface to madwifi. It is easier
to keep this separate to avoid conflicts with potential changes in the
driver interfaces.
TX/RX bytes are now reported correctly (typo ended up leaving TX bytes
uninitialized and set RX bytes value to use correct TX bytes). TX/RX
packet counts are not yet available from kernel, so we have to clear the
values to avoid reporting bogus data.
On WPS init/deinit process, the hostapd clears it's own WPS IE
with 0 length WPS IE. But it fails. Because the parameter to
ioctl is too short. Then hostapd prints a below message.
ioctl[IEEE80211_IOCTL_SET_APPIEBUF]: Invalid argument
Allow more than one pending PutWLANMessage data to be stored (M2/M2D
from multiple external Registrars) and drop pending M2/M2D messages when
the Enrollee replies with M3.
It looks like Intel wsccmd may send a bogus NewWLANEventMAC
(11:22:33:44:55:66) when acting as an wired external Registrar. Work
around this by going through all STAs if the address does not match and
pick the STA that is in an ongoing WPS registration.
If you don't choose OpenSSL as TLS implementation and choose to enable
CONFIG_EAP_TNC you have to link against libdl. The OpenSSL libraries
implicitly link against them, so this might be a reason why it wasn't
noticed yet. I assume the same applies to hostapd.
We can now handle up to 65535 byte result buffer which is the maximum
due to WEXT using 16-bit length field. Previously, this was limited to
32768 bytes in practice even through we tried with 65536 and 131072
buffers which we just truncated into 0 in the 16-bit variable.
This more or less doubles the number of BSSes we can received from scan
results.
MadWifi is unlikely to be in ../head relative to hostapd or
wpa_supplicant, as it would be inside the hostap git repository.
MadWifi sources are more likely to be in a directory called "madwifi"
and residing outside the hostap repository. Using "madwifi" also
demonstrates that the top-level madwifi directory is needed.
Try to match PRI/SEC channel with neighboring 20/40 MHz BSSes per
IEEE 802.11n/D7.0 11.14.3.2. This is not yet complete implementation,
but at least some parts of the 40 MHz coex are improved.
40 MHz operation maybe rejected (i.e., fall back to using 20 MHz) or
pri/sec channels may be switched if needed.
I think that the "radius" pointer in the structure hostapd_config is
never used; when the configuration is parsed the related data is stored
in hostapd_bss_config's "radius" var.
This adds mostly feature complete external Registrar support with the
main missing part being proper support for multiple external Registrars
working at the same time and processing of concurrent registrations when
using an external Registrar.
This code is based on Sony/Saice implementation
(https://www.saice-wpsnfc.bz/) and the changes made by Ted Merrill
(Atheros) to make it more suitable for hostapd design and embedded
systems. Some of the UPnP code is based on Intel's libupnp. Copyrights
and licensing are explained in src/wps/wps_upnp.c in more detail.
The inactivity poll was originally supposed to use Data::Nullfunc, but
due to Prism2/2.5/3 firmware issues, this was changed to an empty
Data::Data frame. mac80211 does not have such an issue, so change the
inactivity poll frame to be Data::Nullfunc by default and use the
Data::Data workaround only with Host AP driver.
Previous version was discarding TX status for FromDS data frames, but
those are the exact ones that we need to check for inactivity poll to
work, i.e., they are TX status reports for injected data frames.
In addition, remove the debug printing of TX status for data frame since
that could fill up the debug output if kernel-side filtering cannot be
used with monitor interface.
TX status information for all transmitted data frames is not going to
be sent to hostapd anymore, so the CPU load with high traffic load is
going to be significantly reduced.
If a Registrar tries to configure the AP, but fails to validate the
device password (AP PIN), lock the AP setup after four failures. This
protects the AP PIN against brute force guessing attacks.
This optional configuration parameter can be used to override AP
Settings attributes in M7 similarly to extra_cred option for Credential
attribute(s) in M8.
Do not initialize EAPOL state machine for the STA when hostapd is
configured to use WPS with open or shared WEP networks. This allows the
STA to use EAPOL-Start to indicate it wants to start WPS in such a case
and hostapd does not end up running through EAPOL authentication timeout
and disconnecting the STA if WPS is not used.
There was already code for starting EAPOL state machines based on
received EAPOL packets, but that was not working properly since
portEnabled was not set to TRUE on that code path. This is now fixed,
too.
This operation can now be moved into an external program by configuring
hostapd with wps_cred_processing=1 and skip_cred_build=1. A new
ctrl_iface message (WPS-REG-SUCCESS <Enrollee MAC addr> <UUID-E>) will
be used to notify external programs of each successful registration and
that can be used as a tricker to move from unconfigured to configured
state.
This behaves like the one in wpa_supplicant, i.e., hostapd can be
configured not to process new credentials (AP settings) internally and
instead pass the WPS attributes for an external program to process over
ctrl_iface.
The default interval is now 5 seconds (used to be 1 second for
interactive mode and 2 seconds for wpa_cli -a). The interval can be
changed with -G<seconds> command line option.
The separate Association Comeback Time IE was removed from IEEE 802.11w
and the Timeout Interval IE (from IEEE 802.11r) is used instead. The
editing on this is still somewhat incomplete in IEEE 802.11w/D7.0, but
still, the use of Timeout Interval IE is the expected mechanism.
This makes it easier to pass the credential data to external programs
(e.g., Network Manager) for processing. The actual use of this data is
not yet included in hostapd/wpa_supplicant.
This cleans up the driver wrapper interface by getting rid of sta_info.h
dependency in all drivers that use MLME implementation in hostapd
(driver_hostap.c and driver_nl80211.c).
driver.h contains the definitions needed in driver wrapper
implementations (driver_*.c) and driver_i.h contains the definitions
that are used in core hostapd code to interact with the driver wrappers.
The configuration parsing functions seemed to have worked fine before,
but these were real bugs even if they did not show up in practice.
hostapd_ip_diff() was broken for IPv6 addresses (overwrote address and
always returned 1.
This updated all doxygen runs to use the same style that was used for
wpa_supplicant full documents. The full vs. fast configurations are now
otherwise identical apart from fast not generating dot files or
latex/pdf version of the documentation.
Generate a SHA1 hash -based UUID from the local MAC address if the UUID
was not configured. This makes it easier to prepare for WPS since there
is no need to generate an UUID.
mac80211 can now figure out which key to use for injected frames (in
most cases), so we can remove the workaround for configuring IGTK on the
monitor interface that is used for injecting frames.
There is not really much else the Authenticator can do if it does not
receive valid EAP response from the Supplicant/EAP peer. EAP-Failure
would need to be sent before trying to start again with
EAP-Request/Identity, but that is not allowed before the EAP peer
actually replies. Anyway, forcing a new association is likely to clean
up peer state, too, so it can help fixing some issues that could have
caused the peer not to be able to reply in the first place.
It looks like this never survived the move from IEEE 802.1X-2001 to
IEEE 802.1X-2004 and EAP state machine (RFC 4137). The retransmission
scheduling and control is now in EAP authenticator and the
calculateTimeout() producedure is used to determine timeout for
retransmission (either dynamic backoff or value from EAP method hint).
The recommended calculations based on SRTT and RTTVAR (RFC 2988) are not
yet implemented since there is no round-trip time measurement available
yet.
This should make EAP authentication much more robust in environments
where initial packets are lost for any reason. If the EAP method does
not provide a hint on timeout, default schedule of 3, 6, 12, 20, 20, 20,
... seconds will be used.
Previously, only the delivery option 1 from RFC 4284
(EAP-Request/Identity from the AP) was supported. Now option 3
(subsequent EAP-Request/Identity from RADIUS server) can also be used
when hostapd is used as a RADIUS server. The eap_user file will need to
have a Phase 1 user entry pointing to Identity method in order for this
to happen (e.g., "* Identity" in the end of the file). The identity hint
is configured in the same was as for AP/Authenticator case (eap_message
in hostapd.conf).
This commit changes just the name and Action category per D7.0. The
retransmit/timeout processing in the AP is not yet updated with the
changes in D7.0.
dot11RSNAConfigGroupUpdateTimeOut and
dot11RSNAConfigPairwiseUpdateTimeOut MIB variables were only used in
draft versions of IEEE 802.11i, so rename these in order not to use
confusing name here.
Replaced EAPOL-Key timeout to use following timeouts (in
milliseconds): 100,1000,1000,1000 (this was 1000,1000,1000,0). There
is no point in sending out the final EAPOL-Key frame which would be
immediately followed by disconnection. After the change to allow
response to any pending EAPOL-Key frame, it is fine to send the first
retransmission quickly to avoid long wait in cases where Supplicant
did not receive the first frame for any reason. The new sequence will
still provide 3.1 seconds of time to get any response frame, so this
does not reduce the previous time.
Accept response to any pending request, not just the last one. This
gives the Supplicant more time to reply since hostapd will now allow up
to three seconds for the reply to the first EAPOL-Key frame transmission
(and two seconds for the first retry and one second for the last) while
the previous version invalidated any old request immediately when
sending a retransmitted frame.
If the Supplicant replies to more than one request, only the first reply
to arrive at the Authenticator will be processed. As far as the
Supplicant is concerned, this behavior does not differ from the previous
one except for being less likely to cause unneeded retransmissions of
EAPOL-Key frames.
This can help in cases where power saving is used when the group key is
rekeyed or when there is excessive traffic on the channel that can delay
(or drop) EAPOL-Key frames.
driver_test can now be used either over UNIX domain socket or UDP
socket. This makes it possible to run the test over network and makes it
easier to port driver_test to Windows.
hostapd configuration: test_socket=UDP:<listen port>
wpa_supplicant configuration: driver_param=test_udp=<dst IP addr>:<port>
Unfortunately, at least the current libnl git snapshot changes the API
in backwards incompatible way and in a way that makes it difficult to
to allow building against the latest libnl code.
This is currently happening way too frequently (mac80211 monitor
interface sends TX info for every frame) and the end result makes it
difficult to read hostapd debug log if there is large amount of data
traffic.
Only one of the authentication frame types is encrypted. In order for
static WEP encryption to work properly (i.e., to not encrypt the frame),
we need to tell mac80211 about the frames that must not be encrypted.
If the phy info from nl80211 does not include 802.11b mode, generate
that mode based on 802.11g information. This allows hw_mode=b to be used
with drivers that support 2.4 GHz band.
This is just making an as-is copy of EAP-AKA server and peer
implementation into a new file and by using the different EAP method
type that is allocated for EAP-AKA' (50). None of the other differences
between EAP-AKA and EAP-AKA' are not yet included.
It is likely that once EAP-AKA' implementation is done and is found to
work correctly, large part of the EAP-AKA and EAP-AKA' code will be
shared. However, it is not reasonable to destabilize EAP-AKA
implementation at this point before it is clearer what the final
differences will be.
Since the Registrar may not yet know the UUID-E when a new PIN is
entered, use of a wildcard PIN that works with any UUID-E can be useful.
Such a PIN will be bound to the first Enrollee trying to use it and it
will be invalidated after the first use.
If a STA reassociates and changes key_mgmt (e.g., from WPA-PSK to WPS),
hostapd needs to reset some of the existing STA and WPA state machine
variables to allow correct processing for the new association.
WPS IE is now passed from hostapd association processing into EAP-WSC
and WPS processing. Request Type attribute is parsed from this
information and if the request is for a WLAN Manager Registrar,
additional management keys are derived (to be used with UPnP).
This depends on a patch to Linux nl80211/mac80211 that has not yet been
merged into wireless-testing. If that change is not present, the old
mechanism (WEXT) will be used instead.
It is better to pass both HT Capabilities and HT Operation IEs in the
same function call since it may be easier for the driver wrappers to
handle the changes without having to wait for the other IE in the
wrapper code.
If country_code is set in hostapd.conf, hostapd will now update nl80211
regulatory data by setting the alpha2 string for CRDA. In other words,
"iw reg set <alpha2>" is not needed anymore when using hostapd.
