Bug #6001
closedosmo-gsm-tester_virtual: netreg_mass fails: Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle
100%
Description
Since March 21 the netreg_mass test in osmo-gsm-tester_virtual has been failing. No MS is performing LU anymore.
22:51:01.677449 run osmo-ms-mob/00098: Starting mobile 22:51:01.704068 run osmo-ms-mob/00098(pid=234): Launched 22:51:01.723323 run mass: MS start registered {at=8155812.010846457, delay=0.008084695786237717, ms=00098} 22:51:03.474475 run osmo-ms-mob/00099: Starting mobile 22:51:03.500497 run osmo-ms-mob/00099(pid=235): Launched 22:51:03.511270 run mass: All started... {duration=55.749718241393566, too_slow=0} 22:51:03.521608 run mass: MS start registered {at=8155813.809176664, delay=0.010319825261831284, ms=00099} 22:52:07.825656 run mass: Tests done {all_completed=False, max=None, min=None} 22:52:07.840405 tst register_default_mass.py:52: ERR: Exception: No run completed. [trial-7152↪netreg_mass↪register_default_mass.py:52] 22:52:07.840904 tst register_default_mass.py:52: Test FAILED (140.7 sec)
Looking at e.g. run.2023-04-11_22-49-48/netreg_mass/osmo-ms-mob/00000/stderr, it seems to be an error in the state machine:
(launched: 2023-04-11_22:50:09.822249) <0013> script_lua.c:118 No handler for Shutdown <0013> script_lua.c:118 No handler for Started <0011> app_mobile.c:254 *** Warning: Mobile '00000' has default IMEI: 000000000000000 <0011> app_mobile.c:256 This could relate your identitiy to other users with default IMEI. *** <0011> app_mobile.c:278 Mobile '00000' initialized, please start phone now! <001b> telnet_interface.c:88 Available via telnet 127.0.0.1 0 <0013> script_lua.c:118 No handler for Started <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle. <0005> gsm48_mm.c:4323 Message MM_EVENT_NO_CELL_FOUND unhandled in state MM idle.
Pau recently worked on related code, assigning to you.
Related issues
Updated by fixeria about 1 year ago
I guess this might be caused by one of my patches:
commit 4949ab7471370bf2d3b63e74c6846340dc638d06 Author: Vadim Yanitskiy <vyanitskiy@sysmocom.de> Date: Thu Mar 16 20:43:18 2023 +0700 layer23/{mobile,modem}: prefix VTY prompt with project's name Change-Id: If7ce89362c75e2c15995dfb08cfcfdcad24e2529
Back in March I had to fix osmo-python-tests.git:
https://gerrit.osmocom.org/c/python/osmo-python-tests/+/32063 InteractVty: re.escape() the VTY prompt in connect()
but osmo-gsm-tester.git has its own implementation of the VTY interface, so I submitted another fix:
https://gerrit.osmocom.org/c/osmo-gsm-tester/+/32281 osmo_vty: re.escape() the VTY prompt in connect()
UPDATE: pespin reports that this does not solve the problem :/
https://jenkins.osmocom.org/jenkins/job/osmo-gsm-tester_virtual-pespin/6/console
Updated by pespin about 1 year ago
- Assignee changed from pespin to fixeria
Regression introduced by osmo-bts.git 8e04613e4f0bf5093e15a6027ac53e3934555708 "osmo-bts-trx: properly activate [CBCH/]BCCH/CCCH". I confirmed it by building a new osmo-bts trial with that patch reverted on top of current master: test then passes.
Updated by pespin about 1 year ago
- Blocks Feature #1572: Circuit Switched Data (CSD) Support in osmo-bts added
Updated by fixeria about 1 year ago
- Status changed from New to Feedback
- % Done changed from 0 to 80
pespin wrote in #note-2:
Regression introduced by osmo-bts.git 8e04613e4f0bf5093e15a6027ac53e3934555708 "osmo-bts-trx: properly activate [CBCH/]BCCH/CCCH". I confirmed it by building a new osmo-bts trial with that patch reverted on top of current master: test then passes.
I reproduced the problem running osmo-bts-virtual locally and submitted a fix:
https://gerrit.osmocom.org/c/osmo-bts/+/32285 osmo-bts-virtual: properly activate [CBCH/]BCCH/CCCH [NEW]
I makes virtphy on my machine "see" broadcast DL traffic again, so I believe it should fix the problem.
Updated by fixeria about 1 year ago
- Status changed from Feedback to Resolved
- % Done changed from 80 to 100
Yup, this patch did the trick. Merged, the job is green again.