Wait for hostapd to enter the PTKINITDONE state before checking
connectivity. This is needed to avoid a race condition with UML
time-travel.
Signed-off-by: Jouni Malinen <j@w1.fi>
It was clearly too easy to get unexpected behavior by accidentially
passing in a string instead of a list of strings to these functions, so
enforce the correct type to notice such issues automatically.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
wait_event() expects a list of events instead of a single event name.
The previous implementation of wait_sta() did not really wait for
AP-STA-CONNECT; instead, it returned the next event from hostapd
regardless of what that event was.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
When going through 4-way handshake, the station side reports
CTRL-EVENT-CONNECTED after having sent out EAPOL-Key msg 4/4. The AP
side reports AP-STA-CONNECT after having completed processing of this
frame. Especially when using UML with time travel, it is possible for
the connectivity test to be started before the AP side has configured
the pairwise TK if the test is triggered based on CTRL-EVENT-CONNECTED
instead of AP-STA-CONNECT.
Add explicit wait for AP-STA-CONNECT in some of these cases to reduce
likelihood of reporting failures for test cases that are actually
behaving as expected. This shows up with "dev1->dev2 unicast data
delivery failed" in the test log.
Do the same before requesting reauthentication from the station side
since that has a similar issue with the EAPOL-Start frame getting
encrypted before the AP is ready for it.
Signed-off-by: Jouni Malinen <j@w1.fi>
There is no actual need for running the authentication server with
driver=nl80211, so simplify this by using driver=none instead. This
frees up apdev[1] for actual AP needs in the test cases.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Use more readable "foo not in bar" construction for the couple of places
that did "not foo in bar".
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This is needed in cases the hexlify() output is used to concatenate with
a string or used in string comparisons.
Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
Signed-off-by: Jouni Malinen <j@w1.fi>
This is a regression test case for FT-over-DS that got broken on
mac80211-based drivers when the extra key reinstallation checks were
added.
Signed-off-by: Jouni Malinen <j@w1.fi>
The generic cmd_execute() function was introduced in a manner that
converted the argument array to a string and used shell to run the
command unconditionally. This is not really desirable, so move back to
using the command array by default and use the single command string
with a shell only when really needed.
Signed-off-by: Jouni Malinen <j@w1.fi>
Add the feature to execute shell commands on each wpa_supplicant/hostapd
interface host. When executing remote tests the interfaces are not all
on a single host so when executing shell commands the test needs to
execute the command on the host which the interface relevant for the
command is on. This patch enables tests to execute the command on the
relevant host.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
The regular hwsim tests use both unicast and broadcast frames to test
the connectivity between 2 interfaces. For real hardware (remote hwsim
tests) the broadcast frames will sometimes not be seen by all connected
stations since they can be in low power mode during DTIM or because
broadcast frames are not ACKed. Use 10 retries for broadcast
connectivity tests for real hardware so that the test will pass if we
successfully received at least one of them.
Signed-off-by: Jonathan Afek <jonathanx.afek@intel.com>
Pass apdev param to hostapd.add_bss(). Kill hardcoded phy param and get
phy base on apdev. These are needed to support operation with a remote
test host.
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com>
We need this for remote host support. From apdev we can get
apdev['hostname'] and apdev['port'].
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com>
This allows the full apdev dict to be passed to the add_ap() function
instead of just ifname. This allows us to handle also remote hosts while
we can check apdev['hostname'], apdev['port']. The old style ifname
argument is still accepted to avoid having to convert all callers in a
single commit.
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com>
At some point, these hostapd_oom_* test cases started to fail with
wpa_msg() allocation failure for the AP-ENABLED event. This resulted in
unnecessary long test execution (waiting 30 seconds for an event that
was dropped). Speed this up by using a shorter timeout.
Signed-off-by: Jouni Malinen <j@w1.fi>
This adds hwsim test ap_vlan_iface_cleanup_multibss. It connects two
stations in different BSS but the same hostapd process. First both
stations are in VLAN 1, then they get reauthenticated into VLAN 2. Due
to the ordering of the stations moving around, this test checks that
bridge and tagged interface referencing counting is done globally, such
that the tagged interface is not removed too early and no bridge is
left over.
Signed-off-by: Michael Braun <michael-dev@fami-braun.de>
This makes it easier to use instances of control interfaces in common
code without having to separately address hostapd and wpa_supplicant
behavior differences.
Signed-off-by: Jouni Malinen <j@w1.fi>
This allows a critical error to be noticed more quickly and reported
more clearly in the test log.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Some kernel debugging options (especially
CONFIG_DEBUG_KOBJECT_RELEASE=y) can add significant blocking time to
interface removal with NL80211_CMD_DEL_INTERFACE and
SIOCGIFBR(BRCTL_DEL_BRIDGE) block for 1-8 seconds. This could result in
the VLAN test cases failing due to a wpaspy timeout on the REMOVE
command even though the issue was only in the kernel debugging code
making the operations significantly slower. Work around this by using a
longer timeout for that control interface command to allow this type of
debug kernel to be used.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
