|Age||Commit message (Collapse)||Author||Files||Lines|
This reverts commit 0a2da132d1c41313a4bb0c5024bc6789531d13bb.
This reverts commit 54cfdfa3396d48c1c09455b98f8d9788d14db27c.
This reverts commit 9e64e8db8367ccd551305e8c0396944a9d044563.
'make deps' always fails if there is no internet uplink.
While trying to add multiple MSCs on SCCPlite, I noticed that the MSC emulation
responds to RESET regardless of the PC. Only respond when the point-codes match
the configured ones.
osmo-msc failed to record the Complete Layer 3 Information LAC and CI in the
MSC-A as well as the VLR record. Since osmo-msc
I194271af2acb37b4f8cc2d106ab2fd2b0d443589, osmo-msc properly records these for
successful Complete Layer 3 procedures.
Incorporate verification of the LAC and CI in all tests calling f_perform_lu()
and f_expect_clear(). Implement by scraping the output of vty
'show subscriber imsi 1234 conn'
Some tests model a failure to attach, or expire the VLR record: for those, add
parameter verify_cell_id to g_pars, and pass it as false, to skip checking the
LAC and CI.
Disable CI checking for all Iu tests globally in f_verify_vty_lac_ci(), see
For the latest build, which does not yet record LAC and CI properly, provide
mp_enable_cell_id_test, which skips all cell id verification if set to false.
Put to effect by docker-playground I052fea208021509e12826c50474b96474e7a58c2.
Depends: Iee1781985fb25b21ce27526c6a3768bf70d4dc9a (osmo-msc)
So far we're sleeping a hole second before verifying the PDCH status of a
dynamic timeslot. That is just to ensure that osmo-bsc has read the message
from its queue and changed its status.
Change to a sleep of 0.1s, cutting a couple of seconds off the test execution.
Store all System Information data received during RSL bootstrap in a list, then
verify that the 12 expected SI are indeed sent.
The SI data presence and content are recorded, but not yet verified, leaving
that to a future patch.
At the time of writing Ief0d9b096feeee7d37b5f2429dd3e80de0161806 I wasn't aware
of the 'inout' keyword, which allows to pass the counter list by reference.
Rather modify the counter lists in-place. Instead of requiring
list := f_counter_name_vals_add(list, ...)
rather implement by directly modifying list:
Use new f_counter_* functions to verify osmo-bsc MSC pooling counters.
This nicely also verifies the intended effect of each test in detail.
Depends: I2ded757958dfa62b502efbab765203bcadf899e2 (osmo-bsc)
First user will be new MSC pooling tests in ttcn3-bsc-test, see
Some tests may expect that behavior so let the test decide what to do
Not all osmo-bts backends do support multiple transceivers, while
we still want to run test cases against them. Let's make the
number of transceivers configurable (mp_transceiver_num), so it
can be adjusted depending on osmo-bts backend to be used.
For more details, see I25e10e28de174337233e6a3bb32cc16f2d7d614e.
Since change  has been merged, we see multiple regressions in
ttcn3-bsc-test (all LCLS test cases) and ttcn3-bsc-test-sccplite
(sporadic failures). In all failed cases, the reason is similar:
RSL for unknown Dchan
BSC_Tests.ttcn:4501 BSC_Tests control part
BSC_Tests.ttcn:2176 TC_assignment_codec_fr testcase
The mentioned change enables TCP_NODELAY option for all IPA based
connections, including both OML and RSL. This option disables
Nagle's algorithm , so we get less delays on IPA based links.
It took me a lot of time to investigate, and finally, I figured
out what is actually causing those regressions. The TCP_NODELAY
itself is not a problem, of course. As it turned out, the
problem is here, in our TTCN-3 test case framework.
Each test case involves several components (actors) running in parallel.
One of them is RSL_Emulation_CT, which is responsible for handling and
routing of RSL messages between the connected components.
A test case may register dedicated channel handlers by calling
f_rslem_register(), so DCHAN/RLL/IPACCESS messages will be matched
by RslChannelNr/TrxNr and routed to the corresponding one.
If no handler is found for a given RSL message, the RSL_Emulation_CT
would abort the test case execution. And that's where the problem is.
Given that all components are running in parallel, it may happen
that a received RSL message would be processed by the RSL emulation
component faster than the test case would call f_rslem_register().
The test case would be aborted due to "RSL for unknown Dchan".
Speaking in context of the failing BSC test cases, a test case
calls f_rslem_register() on receipt of an Assignment Command as
it contains all the assignment parameters. After that we expect
to receive an RSL ip.access CRCX for that channel.
The problem is that both Assignment Command and ip.access CRCX
messages are sent by the BSC simultaneously, so the later may
be handled faster than the first one. Race condition!
Let's work this around by maintaining a waiting queue, where the
messages, for which no handler was found, will be kept until the
corresponding dedicated channel is registered.
This is an optional feature that needs to be enabled explicitly
by calling f_rslem_dchan_queue_enable(), and then explicitly
disabled by calling f_rslem_dchan_queue_disable().
If at the moment of calling f_rslem_dchan_queue_disable() the
waiting queue is not empty, e.g. because the IUT sent us more
messages than we expected, test execution will be terminated.
The actial fix for the LCLS test cases will be submitted next.
It appears some changes were made only to the files in
docker-playground.git, but not here. This means that running
tests locally produced unexpected results.
We must always ensure that the tests run both without and with
docker, which means making sure the configs are all maintained!
This was removed when libfilter was removed from osmo-bsc.
For some strange reason apparently the config files were not used/tested
while testing the removal of libfilter.
Also, irrespective of breaking TTCN3 test execution, we must introduce
a dummy 'access-list' VTY command to osmo-bsc to ensure we don't break
virtually every config file out there.
They perform several handover runs and tend to timeout prematurely on my
The TC_ho_neighbor_config_* tests sometimes take longer than 30 seconds,
because they run multiple handovers. Since they don't have access to the
Test_CT, they cannot restart the T_guard. The simplest solution is to choose a
longer T_guard timeout for those tests specifically, by adding an argument to
f_init(). (A longer timeout for those tests is following in another patch.)
Why f_init()? Assigning a different default value to T_guard seems to not be
possible, but a different timeout value can be passed to T_guard.start(), which
happens in f_init().
Handover testing required passing MSC and BSC addresses to f_tc_* functions and
added pars.handover.sccp_addr_msc and .handover.sccp_addr_bsc.
MSC pool tests added a separate sub-record pars.mscpool which also contains
these two fields.
Move them both up one level, to form a single pair of pars.sccp_addr_msc and
This eliminates the pars.handover sub-record.
The MSC pooling feature is implemented in osmo-bsc
A VTY command ('mscpool roundrobin next') that allows deterministic testing is
added in I2155d906505a26744966f442ffb1e87a6a9b494c.
osmo-bsc.cfg changes needed for these tests to succeed are in docker-playground
The new tests will fail until the above have been merged.
Let's test whether paging is triggered after T3314 expiration, as well
as suspend and explicit and implicit resume work as expected
In some scenarios we do not expect a PDU from GTP to show up on
the Gb side. Let's add support for this.
So far, the ConnHdlr components were only able to exchange BSSGP PDUs
on the PTP BVC, but not on the SIGNALING BVC. For this, we need to
actually connect to the BSSGP_SP_SIG that BSSGP_Emulation provides.
Add BSSAP_N_UNITDATA_req to RAN_Conn_PT, so that we are able to send a Paging
from a test function that runs on MSC_ConnHdlr.
This will be needed by upcoming MSC pool tests, see
For tr_RSL_PAGING_CMD, also check comp_ref against null.
Upon receiving a tr_RSL_PAGING_CMD, the code tries to dispatch the Paging
Command to all valid clients. However, the ConnectionTable[*].comp_ref is
*always* present, and actually null for unconnected clients.
So, before this patch, a Paging Command from osmo-bsc gets sent to a null
client, which disconnects the RSL emulation and aborts the test.
There is currently no test using this, but the upcoming MSC pool tests will:
Change I16cc50dfca102030380a06e16c234d5f6698f38f fixes them both.
Unlike the RSL_IE_MS_Power, where power_level is 5 bit long, in
the RSL_IE_BS_Power it's 4 bit long. Fix this.
Similar to the MSC tests, have several g_bssap and mp_bssap_cfg.
Prepare for MSC pool tests.
Replace g_bssap with a g_bssap[NUM_MSC] array.
Replace mp_bssap_cfg with an mp_bssap_cfg[NUM_MSC] array.
Requires patch I1986e4ef43beee161c82193694421b56136c1afe in docker-playground
to match the new required BSC_Tests.cfg format.
So far used only in MSC_Tests.ttcn, but soon to be used also in BSC_Tests.ttcn.
f_gen_imei() calls f_enc_IMEI_L3() with a 14 digits argument, but the IMEI_L3
template used is hardcoded to 15 digits. So the oddevenIndicator must always
indicate odd, not depend on the digits argument.
f_gen_imei() should probably also compose a Luhn checksum, leaving that to
Found by using the new osmo_mobile_identity API in osmo-msc, which is stricter
about odd/even and filler digits than our previous implementations.
See osmo-msc Idfc8e576e10756aeaacf5569f6178068313eb7ea .
There are some M3UA implementations out there who use a routing context
during the ASPAC procedure, but who then don't use it in subsequent DATA
transmission. Let's add a test case for that.
Another surprise from the latest osmo-bts release: it may send us
CCCH LOAD INDication message during the RSL bring up. Ignore it.
Unfortunately, the latest release of osmo-bts still has a bug,
that has been fixed  in the recent master. Because of that,
most of the test cases in ttcn3-bts-test-latest currently fail.
The problem is that all transceivers use IPAC_PROTO_RSL_TRX0,
regardless of what the BSC tells them to use. Let's work this
around by patching IPA stream ID in ASP_RSL_Unitdata messages
coming from the IPA emulation.
This test verifies power ramping (up) is working fine during BTS
config files are updated to make sense:
* "nominal power" in osmo-bsc.cfg reflects correct default nominal tx
power of fake_trx.
* "osmotrx tx-attenuation" in osmo-bts.cfg is removed to let osmo-bts
use the value received through OML (max_power_red 20).
* "power-ramp step-size" in osmo-bts.cfg is increased to speed up the
test. There's no good reason to keep it lower.