Flush scan results to avoid failure caused by incorrect channel
selection based on an old result for the same BSSID. This was found with
the following test sequence:
ap_track_sta_no_auth dpp_network_intro_version_missing_req dpp_controller_relay_pkex
Signed-off-by: Jouni Malinen <j@w1.fi>
The first sock.recv() may return both the status,RUNNING and the
following status line if the sigma_dut process ends up being faster in
writing the result than the test script is in reading the result. This
resulted in unexpected behavior and odd error messages when parsing the
result in the test cases. Fix this by dropping the status,RUNNING line
from the result in case the buffer includes multiple lines.
Signed-off-by: Jouni Malinen <j@w1.fi>
This test case was assuming the Configurator would change the
netAccessKey curve every time based on the protocol keys, but that is
not the case anymore, so force that change here for a negative test.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This test case was assuming the Configurator would change the
netAccessKey curve every time based on the protocol keys, but that is
not the case anymore, so force that change here for a negative test.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Do not leave a large number of network added/removed events remaining
for the following test case to handle. This removes some possible
failure test case sequences like the following one:
wpas_ctrl_many_networks dbus_ap_scan_2_ap_mode_scan
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This test cases can fail if previously executed tests leave older scan
results in cfg80211 scan table. Clear that scan table explicitly to
avoid such issues.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This is needed to avoid pri/sec channel switching based on potential
scan results from the previous test cases.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
min-seq.py can be used to find a minimal test sequence that can be used
to reproduce test failures. This is meant for being able to process the
recently added "Failure sequence:" entries from parallel-vm.log to
reduce manual work needed to debug commonly failing test case sequences.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This is needed to avoid test failures when a previous test case might
have restricted the set of allowed SAE groups.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Flush scan results to avoid unexpected behavior due to scan results
remaining available from previous test cases.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Flush scan results to avoid unexpected behavior due to scan results
remaining available from previous test cases.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
The test case name ap_ft_pmf_over_ds was used for two different test
cases which resulted in only one of those being used. Fix this by using
unique test case names.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Clear scan cache to avoid issues with old scan results from earlier test
cases. This caused issues like the following test case sequence failing:
rrm_beacon_req_active_ap_channels ap_ft_eap_dis_over_ds
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Clear scan results to avoid issues with get_bss() finding an entry from
an earlier test case when checking for mesh information.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Clear the scan cache on the AP before running this test since the HT40
operation on the 2.4 GHz band might get disallowed based on scan results
from earlier test cases. This was found with the following hwsim test
case sequence failing:
ap_acs_with_fallback_to_20 wpa2_ocv_ap_ht_mismatch
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Flush the scan table explicitly to avoid issues with the ROAM command if
the new AP is not found and an entry from a previous test case is used
instead. This was happening in a number of cases where a SAE test case
was run after sigma_dut_ap_cipher_gcmp_256 which used the second AP
instance and allowed that to show up in the scan results in the next
text case.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
These can fail if the scan results from the previous test case remain,
e.g., when run immediately after scan_bss_limit.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Add a couple of more channel configuration cases and log the channel
parameters with more details in the test log.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
A number of test cases using 40 MHz or wider channels with the primary
channel 36 were failing when executed after dpp_chirp_ap_5g since that
test case was running an AP on the channel 40 and resulting in need to
swap the primary and the secondary channels in the following test case.
Fix this by clearing the AP scan cache explicitly for such cases.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Flush the scan cache for all test cases that used get_bss() to check for
particular ANQP information. This was already done for one such case
based on commit dd900637b2 ("tests: Make gas_anqp_extra_elements more
robust"), but other test cases need this as well.
This was showing with frequent errors in test cases sequences like this
one:
dfs_radar_no_ht gas_fragment_with_comeback_delay gas_unknown_adv_proto gas_anqp_venue_url
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This is needed to avoid leaving behind a BSS entry with WPS enabled for
the next text case in some cases. In particular, this was causing issues
in the following sequence of test cases:
ap_wps_conf_chan14 ap_wps_cancel ap_wps_pin_request_file
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Now that wpa_supplicant does this internally as a part of the FLUSH
command, there is no need for the test scripts to try to clear the
parameter between test cases.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This test case could fail in some sequences like "sigma_dut_sae
scan_parsing" due to the ignore_old_scan_res parameter accidentally
being left to 1 by the former test case and the simulated scan result
being older than the previous scan trigger. Reduce the age of that scan
entry to make this less likely to happen.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
The DPP Enrollee might wait for up to 60 seconds for the configuration,
so use a longer timeout value to be able to cover this negative test
case where the Configurator never sends the response.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
This group should not be used with SAE and as such, it could cause
confusing test errors here. Use an acceptable group instead.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
At least for the time being PKEXv2 needs CONFIG_DPP3=y to work in a
testable manner. Couple of the test cases did not cover this correctly
and resulted in failures (instead of skipping the tests) when the
default build configuration was used. Fix that by checking for DPP
version 3.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
Before, we could loose some events because of pipe buffering. I saw this
problem when running "ubus listen" or "logread -f" and waiting some
specific events. After disabling buffering this works much better.
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@gmail.com>
This parameter was added with the commit message indicating the valid
reason code values to be 1-5, but with the implementation allowed only
1. There are five defined reason code values for the Association
Disallowed attribute, so extend the allowed range to cover all those
values.
Fixes: fb9a1c3e28 ("hostapd: Add MBO IE to Beacon, Probe Response, Association Response")
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@gmail.com>
Add the "InterworkingSelect" method to the DBus API to trigger an
Interworking scan with ANQP fetches. When a BSS that matches a
configured credential is found, the result is emitted using the signal
"InterworkingAPAdded". Completion of the full InterworkingSelect
operation is indicated with the "InterworkingSelectDone" signal.
Signed-off-by: Damien Dejean <damiendejean@chromium.org>
Add "AddCred", "RemoveCred", and "RemoveAllCreds" methods to the D-Bus
API of the network interface to allow the caller to manipulate a set of
Interworking credentials similarly to the way this was enabled through
the control interface.
Signed-off-by: Damien Dejean <damiendejean@chromium.org>
CONFIG_DPP3=y can now be used to configure hostapd and wpa_supplicant
builds to include DPP version 3 functionality. This functionality is
still under design and the implementation is experimental and not
suitable to be enabled in production uses before the specification has
been finalized.
Signed-off-by: Jouni Malinen <quic_jouni@quicinc.com>
When checking DPP capabilities the Brainpool flag was not always set
when needed, leading to run a test with the Brainpool curves not
supported by BoringSSL.
Use a short form for the DER length of EC privateKey with NIST P-521
curve. Indeed BoringSSL returns an error when parsing DER sequence 30 81
50 ... because the length 81 50 could have been encoded as 50 and
according comment in BoringSSL:
ITU-T X.690 section 10.1 (DER length forms) requires encoding the
length with the minimum number of octets.
Signed-off-by: Cedric Izoard <cedric.izoard@ceva-dsp.com>
As BoringSSL version of i2d_PUBKEY() doesn't respect the
POINT_CONVERSION_COMPRESSED flag redefine a specific
crypto_ec_key_get_subject_public_key() version for BoringSSL based on
dpp_bootstrap_key_der().
The only other user of crypto_ec_key_get_subject_public_key() is SAE-PK
for which the public key should also be formatted using compressed
format.
Signed-off-by: Cedric Izoard <cedric.izoard@ceva-dsp.com>
Move code of dpp_get_pubkey_point() to a crypto library specific
function crypto_ec_key_get_pubkey_point().
Signed-off-by: Cedric Izoard <cedric.izoard@ceva-dsp.com>
Add a common well-known regulatory database to the test VMs during runs
to remove one thing to have correct in the host.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Replaced the word "sanity" with the inclusive word "validity". The
comment in acs_survey_interference_factor() was referring a function
that does not exist, so remove it instead of trying rename the function.
Signed-off-by: Arowa Suliman <arowa@chromium.org>
Using __del__ for any kind of cleanup is not a good idea
as it's not guaranteed to be called at any particular time,
it's only called whenever the next garbage collect cycle
kicks in.
Use a context manager instead, which basically removes the
need for the try/finally and fixes the reliance on __del__.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Couple of the PASN test cases did not verify whether the wpa_supplicant
build used in the test included PASN support.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
This test case adds a new AP device (wlan0_ap) with iw and removes it in
the end. However, the hostapd interface for this netdev was only added,
but not removed at the end of the test case. This could result in
consecutive test cases getting confused with the extra interface, e.g.,
if running WPS configuration steps that get applied to all enabled
interfaces.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Need to wrap back to 0 when changing value 255 to avoid generating a too
large value to fit an octet field. This was resulting in errors due to a
python exception (likely for about every 256th run).
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
The previous versions expired, so need to re-sign these to fix number of
the EAP test cases. This contains updates from running
tests/hwsim/auth_server/update.sh.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Move from RS to PA country code to allow these test cases to work since
regdb was updated to require DFS for these operating classes in RS.
Signed-off-by: Jouni Malinen <j@w1.fi>
Add minimal testing for airtime policy configuration. mac80211_hwsim
does not actually support this functionality, so this is just for
testing coverage of src/ap/airtime_policy.c.
Signed-off-by: Jouni Malinen <j@w1.fi>
Dump pending monitor interface messages between each roaming step to
make the test log easier to understand and hostapd wait for the new
connection more robust by ensuring that the processed event if for the
very last reassociation. It looks like at least ap_ft_vlan_over_ds_many
could fail due to the connectivity check being started before the final
roam had been completed on the AP side even though there was an explicit
hapd2ap.wait_sta() wait before the test.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
APs PMF capabilities can differ. wpa_supplicant should be able to
disable and enable MBO when roaming to and from a misbehaving MBO AP
that doesn't support PMF. Verify that this is indeed happening.
Signed-off-by: Matthew Wang <matthewmwang@chromium.org>
gcc-10 seems to be inlining eap_eke_prf() and eap_eke_prfplus() which
breaks this test case due to a different backtrace being generated for
triggering the local failures. Point to the functions called by those
instead of these two functions to get this working with both gcc-9 and
gcc-10.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Hide /usr/share/wireshark from hostfs to prevent tshark from loading all
the data from there since that can take significant amount of time and
is not really needed for the test cases. In addition, set HOME to point
to local tmpfs to avoid unnecessary references through hostfs.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Even though the STA in this test case does not actually use SAE, it
needs to recognize the "SAE H2E only "BSS membership selector.
Signed-off-by: Jouni Malinen <j@w1.fi>
Need to explicitly wait for hostapd to report STA connection before
starting the traffic test to avoid the potential race condition when
testing with UML and time travel mode.
Signed-off-by: Jouni Malinen <j@w1.fi>
These could fail if a scan entry from a previous test case was still
present in the BSS table, e.g., by wpa_supplicant selecting the SSID
from that old entry instead of the new SSID. Try to avoid that by
explicitly flushing the scan results before starting these tests.
Signed-off-by: Jouni Malinen <j@w1.fi>
Need to close the WpaSupplicant instance on the extra radio before
returning from this test case since that interface is going to be
removed and WpaSupplicant.__del__() can time out on trying to detach the
monitor connection after that.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Make it more difficult to miss issues that were previously only printed
out in /tmp/hwsim-test-logs/*-parallel.log. This covers things like
memory leaks and test script failures or forgotten development time
prints to stdout.
Signed-off-by: Jouni Malinen <j@w1.fi>
This test case was missing an explicit CAPI ap_reset_default and that
could result in hostapd being left running at the end of the test case.
This could result in issues with following test cases if they used a new
radio interface from HWSimRadio().
Signed-off-by: Jouni Malinen <j@w1.fi>
Raising an exception while the wlan5 interface was remove (i.e., between
wpas.interface_remove() and .interface_add() calls) would result in the
cleanup code failing and generating yet another exception while the
first one was being processed. Work around this by re-adding the wlan5
interface back temporarily if the interface is not available for the
cleanup operations.
Signed-off-by: Jouni Malinen <j@w1.fi>
These commands were being issues to incorrect wpa_supplicant instance
and were missing clearing of the MAC_RAND_SCAN parameter.
Signed-off-by: Jouni Malinen <j@w1.fi>
The iteration of WpaSupplicant instances used incorrect variable and
ended up cleaning up only the wlan5 interface. This left unexpected
setband parameter for wlan0/wlan1/wlan2 which could result in
consecutive test cases failing due to scan not finding the expected
BSSs.
Signed-off-by: Jouni Malinen <j@w1.fi>
It was possible for these test cases to fail if the first scan iteration
did not find the AP since the 10 second timeout was small enough to
terminate the second attempt before fetching the scan results. Increase
this timeout to allow at least two full scan iterations to be completed
before declaring failure.
Signed-off-by: Jouni Malinen <j@w1.fi>
These test cases use hidden SSIDs and left behind a BSS entry with no
SSID. That can cause issues for consecutive test cases where the BSSID
can be used as the key for finding a BSS entry. That could end up
picking the old hidden SSID BSS instead of the one that was meant to be
used in the test case.
Flush the scan cache at the end of the scan-ssid-list test cases to
reduce invalid test failures for the consecutive test cases.
Signed-off-by: Jouni Malinen <j@w1.fi>
The PASN_START command can fail if there is an old BSS entry for the
same BSSID from an earlier test case. Try to avoid this by flushing the
scan results before running these test cases.
Signed-off-by: Jouni Malinen <j@w1.fi>
The isAlive() function is deprecated in newer versions of Python
so replace it with the is_alive() instead.
Signed-off-by: Oren Givon <oren.givon@intel.com>
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Need to clear sae_groups parameter before using SAE in this test case to
avoid issues if previous test cases have left a specific group
configured.
Signed-off-by: Jouni Malinen <j@w1.fi>
This could fail in theory if running out of memory, so better check for
this explicitly instead of allowing the exchange to continue and fail
later due to checkcode mismatch.
Signed-off-by: Jouni Malinen <j@w1.fi>
Add test cases that check preCAC, which is available for EU regulatory
domain. Also confirm that preCAC is not used for US.
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@gmail.com>
Add option to:
- add a new AP on the same phy that the backhaul-sta uses
- run CSA from the parent
Adding a new AP (backhaul/fronthaul) on the same phy we have for
backhaul-sta is closer to the real repeater implementation.
Add a test case for that and run CSA.
This is a common problem when we have on the same phy:
- connected backhaul STA
- we started fronthaul/backhaul AP
- we receive (from parent) CSA on the STA interface
This is multi_ap_wps_shared_apdev_csa test case, which fails today with
both mac80211_hwsim and ath9k. To avoid always failing test cases,
ignore this failure for now. Full validation can be enabled once the
issue behind this is fixed.
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@gmail.com>
Pass the backhaul parameters as a parameter. This is in preparation for
channel switch test for Multi AP.
Signed-off-by: Janusz Dziedzic <janusz.dziedzic@gmail.com>
This test first configure hostapd with an initial SSID
('test-wpa2-psk-start'). Then a new SSID is configured
('test-wpa2-psk-new') using SET and RELOAD. Next, a station is
associated using WPS, and the test verifies that the new SSID was served
to the station.
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
cfg80211 may ignore user hints while there are active COUNTRY_IE hints,
thus at some timings it may ignore the country setting back to world
domain. Fix it by making sure the country is set only after all the
interfaces are stopped. In addition, call a more robust
clear_regdom_dev() function.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Test that if notify_mgmt_frames is enabled and a station connects we do
get AP-MGMT-FRAME-RECEIVED, and that it includes an Authentication
frame.
Also test that if notify_mgmt_frames is disabled, no Management frame is
sent on ctrl_iface when a station connects.
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
The P2P group may be originally formed on UNII-3, so disabling UNII-1
and UNII-2 will not result in a channel switch failing the test.
Fix this by setting 44 as a preferred channel.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
The test configures ft_r0_key_lifetime parameter, however ft_params
already contain the r0_key_lifetime. Since both options are accepted by
hostapd and set the same field, one of them gets overwritten.
As the dictionary enumeration order is not guaranteed in python, the
test may sporadically fail.
Fix that by explicitely removing the unneeded parameter.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Kernel commit 14486c82612a ("rfkill: add a reason to the HW rfkill
state") added an extra byte to the end of the rfkill events and that
confused the read loop here since python tried to buffer the results
from multiple read() calls into the local buffer that then delivered the
extra octets in consecutive events. Fix this by disabling buffering for
these reads.
Signed-off-by: Jouni Malinen <jouni@codeaurora.org>
Python enumerates dictionaries in an arbitrary order, thus the resulting
configuration file will have the parameters shuffled randomly. This may
cause the test to fail when, for example, auth_server_addr is placed
after auth_server_port. Fix this by enforcing some of the fields to be
placed before the others.
Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
