OpenBSC

openbsc@lists.osmocom.org

4317 discussions

New repository - libosmo-gprs.git
by Vadim Yanitskiy 18 Aug '22

18 Aug '22

Hi all, as some of you may know, I am working on the MS side implementation of GPRS protocol stack. The end goal is to have something similar to srsUE (part of srsRAN, former srsLTE), but for GPRS. In order to reduce code duplication it was decided to move some stuff from both osmo-pcu.git and osmo-sgsn.git into shared libraries, so that we could re-use existing code in the upcoming MS side implementation. == What is already done * Redmine: https://osmocom.org/projects/libosmo-gprs * Gerrit: https://gerrit.osmocom.org/q/libosmo-gprs * Gitea: https://gitea.osmocom.org/osmocom/libosmo-gprs * osmo-ci.git: https://gerrit.osmocom.org/c/osmo-ci/+/29019 == What is missing * GitHub: https://github.com/osmocom/libosmo-gprs * cgit: https://cgit.osmocom.org/libosmo-gprs/ ** Coverity pulls from this mirror ** Jenkins build verification uses this mirror I will need some help with both GitHub and cgit, as I don't have admin permissions. Please also let me know if I missed something. Best regards, Vadim. -- - Vadim Yanitskiy <vyanitskiy at sysmocom.de> http://www.sysmocom.de/ ======================================================================= * sysmocom - systems for mobile communications GmbH * Alt-Moabit 93 * 10559 Berlin, Germany * Sitz / Registered office: Berlin, HRB 134158 B * Geschaeftsfuehrer / Managing Director: Harald Welte

4 8

shadysim STK installation issue
by Karsten Ohme 11 Aug '22

11 Aug '22

I tried to use shadysim (https://gitea.osmocom.org/sim-card/sim-tools) to install a STK / CAT applet on the sysmoISIM. It was not working for me, also it does not support UICC parameters and it seems that still Python 2.7 is needed. As an alternative I have added support for gpshell (https://kaoh.github.io/globalplatform/) to use SCP02 and pass the SIM/UICC parameters. See the new parameters "uiccSystemSpecParam" and "simSpecParam" for the "install" command https://github.com/kaoh/globalplatform/blob/master/gpshell/src/gpshell.1.md and inspect the https://github.com/kaoh/globalplatform/blob/master/gpshell/helloInstallSTK.… example file. If "brew" ins installed then "brew install kaoh/globalplatform/globalplatform" is sufficient for the installation. I also faced the issue that the sysmoISIM cards are only accepting SIM parameters and no UICC parameters, which is strange or even a bug since these cards are newer and are supporting the newer UICC ETSI specifications.

1 0

Question about SDR aix-tx project
by zhang_xiangjun 20 Jul '22

20 Jul '22

Dear Sir/Madam， I'm a graduate student from a martime college , my major is not computer or communication , I'm trying to do a project about vessel's AIS tx . The equipment which I use is USRP b210 and the software is Gnuradio . After I cloned the code from http://github.com/osmocom/ais-tx/tree/master/gr-aistx and install into gnuradio , it occurs a few mistakes . I tryed to install it in Ubnutu system and dragonOS system ,both failed . Could take a look and help me to solve the problem , there is some pictures attached in this e-mail about the problem . Really thanks a lot , wish you have a nice day . Yours Sincerelly Xiangjun Zhang

1 0

Operating multiple MSs within one instance of mobile
by James Connor 13 Jul '22

13 Jul '22

Hey y'all, This is somewhat related to another post of mine titled "MSC, SIP connector, and Asterisk". I've been working on a project to simulate cell phone usage through construction of a completely virtual GSM network. To that end, I've been planning to use the mobile application packaged with OsmocomBB to simulate the actual MSs. To make this easier, I've been using a single instance of mobile configured with several MSs in the configuration file (alongside separate instances of virtphy). This had been working for a while, where I was able to place calls between the MSs that were configured. However, I've recently run into an issue where the calls are rejecting each other. After digging through the source of mobile a little, it's looking increasingly like multiple MSs are not meant to be used in parallel with each other on the same instance of mobile. The main file I've been looking at is mnccms.c (https://osmocom.org/projects/baseband/repository/osmocombb/revisions/master…). It seems that there's no differentiation made between MSs when checking for any calls with/without hold status, which means that only one active call can be used by a mobile instance at any one time. In fact, based on what I've been reading, I can't figure out how I had this working in the first place. Therefore, I wanted to see if I could get a definitive answer before I invest time into redesigning certain aspects of my framework. Is it possible to use multiple MSs within a single instance of mobile, such that they can all be involved in separate calls simultaneously? Thanks for any help, it's greatly appreciated! - James

1 0

MSC, SIP connector, and Asterisk
by James Connor 11 Jul '22

11 Jul '22

Hey all, I've got a bit of an interesting use case. I've been working to setup the OpenBSC components in order to simulate stress testing a GSM network. I've been able to get communications across the GSM working properly, with MSs able to place calls. However, part of what I'm trying to do involves routing to a simulated PSTN. I turned to sip-connector and asterisk, but my network immediately stopped working once I told the MSC to use an external MNCC. The setup involves two separate computers. The first is the BSS, which has a BSC, MGW, virtual BTS, and two MSs running on mobile and virtphy. The second is the NSS, with the MSC/VLR, HLR, MGW, and STP. The GSM is split between the two devices because of the eventual goal to increase the number of BSSs to better simulate a real-world GSM. This was all working fine, until I modified osmo-msc.cfg to include "mncc external /tmp/msc_mncc". In parallel, I setup osmo-sip-connector and ran it on the NSS through the same "/tmp/msc_mncc" socket, pointing it at Asterisk (also on the NSS). This was all done according to the instructions listed here: https://osmocom.org/projects/osmo-sip-conector/wiki/Howto. Now whenever I place my calls (which I do through mobile's "call" VTY command on the BSS), the call doesn't connect and simply gets released after 30 seconds. After doing some debugging, I've determined that Asterisk is not receiving anything from the sip-connector. Meanwhile, the sip-connector is being told to delete any connection basically as soon as it's created: <0001> mncc.c:1005 MNCC rcvd message type: MNCC_SETUP_IND <0001> mncc.c:566 Created call(5004) with MNCC leg(2147483652) IMSI(001010000000001) <0001> mncc.c:68 Starting Timer for MNCC_RTP_CREATE <0001> mncc.c:164 MNCC sent message type: MNCC_RTP_CREATE <0001> mncc.c:1005 MNCC rcvd message type: MNCC_REL_IND <0001> mncc.c:636 Rcvd MNCC_REL_IND, Cause: RESOURCE_UNAVAIL <0001> mncc.c:648 leg(2147483652) was released. <0002> call.c:90 call(5004) released. The MSC has the following as part of its output when a call is placed: ... <0020> osmo_ss7.c:1933 0: asp-OsmoMSC-asp: xua_cli_read_cb(): sctp_recvmsg() returned 48 (flags=0x80) <0023> m3ua.c:714 0: asp-OsmoMSC-asp: Received M3UA Message (XFER:DATA) <0023> m3ua.c:543 0: asp-OsmoMSC-asp: m3ua_rx_xfer <0023> m3ua.c:566 0: asp-OsmoMSC-asp: m3ua_rx_xfer(): M3UA data header: opc=2233=1.23.1 dpc=185=0.23.1 <0020> osmo_ss7_hmrt.c:276 m3ua_hmdc_rx_from_l2(): found dpc=185=0.23.1 as local <0020> sccp_scrc.c:472 scrc_rx_mtp_xfer_ind_xua: HDR=(CO:CODT,V=0,LEN=0), PART(T=Destination Reference,L=4,D=00000007), PART(T=Segmentation,L=4,D=00000000), PART(T=Data,L=6,D=000422040120) <0021> sccp_scoc.c:1664 Received CO:CODT for local reference 7 <0021> sccp_scoc.c:1698 SCCP-SCOC(7)[0x555adb5ddf40]{ACTIVE}: Received Event RCOC-DT1.ind <0021> sccp_user.c:175 Delivering N-DATA.indication to SCCP User 'OsmoMSC-A' <0011> sccp_ran.c:108 (GERAN-A-7) sccp_ran_sap_up(N-DATA.indication) <0003> ran_peer.c:591 ran_peer(GERAN-A:RI-SSN_PC:PC-1-23-1:SSN-BSSAP)[0x555adb5b1010]{READY}: Received Event RAN_PEER_EV_MSG_UP_CO <0007> ran_peer.c:407 msc_i(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5d97e0]{READY}: Received Event MSC_EV_FROM_RAN_UP_L2 <0011> ran_msg_a.c:790 msc_i(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5d97e0]{READY}: RAN decode: BSSMAP: CLEAR REQUEST <0007> msc_i.c:85 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_COMMUNICATING}: Received Event MSC_A_EV_FROM_I_PROCESS_ACCESS_SIGNALLING_REQUEST <000b> msc_a.c:196 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_COMMUNICATING}: + msc_a_ran_dec: now used by 2 (cc,msc_a_ran_dec) <0011> ran_msg_a.c:790 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_COMMUNICATING}: RAN decode: BSSMAP: CLEAR REQUEST <0011> msc_a.c:1612 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_COMMUNICATING}: RAN decode: BSSMAP Clear Request <0007> msc_a.c:1433 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_COMMUNICATING}: Received Event MSC_A_EV_MO_CLOSE <0007> msc_a.c:733 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_COMMUNICATING}: State change to MSC_A_ST_RELEASING (X2, 30s) <0011> msc_a.c:769 msc_a(IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ)[0x555adb5daa50]{MSC_A_ST_RELEASING}: Releasing: msc_a use is 2 (cc,msc_a_ran_dec) <000b> msc_a.c:773 VLR subscr IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97 + msc_a_fsm_releasing_onenter: now used by 4 (attached,active-conn,CC,msc_a_fsm_releasing_onenter) <000b> vlr.c:309 VLR subscr IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97 + vlr_subscr_cancel_attach_fsm: now used by 5 (attached,active-conn,CC,msc_a_fsm_releasing_onenter,vlr_subscr_cancel_attach_fsm) <000b> vlr.c:314 VLR subscr IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97 - vlr_subscr_cancel_attach_fsm: now used by 4 (attached,active-conn,CC,msc_a_fsm_releasing_onenter) <0001> transaction.c:230 trans(CC:INITIATED IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ callref-0x80000003 tid-8) Freeing transaction <0005> gsm_04_08_cc.c:237 trans(CC:INITIATED IMSI-001010000000001:MSISDN-1234567891:TMSI-0x9B96EF97:GERAN-A-7:CM_SERVICE_REQ callref-0x80000003 tid-8) tx MNCC_REL_IND ... Any thoughts as to why this one change to my GSM is suddenly stopping it from functioning properly? Thanks, any help is greatly appreciated! - James P.S. I didn't want to include my config files for fear of making this message absurdly long, but let me know if you want to see any and I'll gladly post them.

4 6

Interfacing Osmocom GSM network to USA PSTN
by Mychaela Falconia 02 Jul '22

02 Jul '22

Hello Osmocom community, I have previously alluded in other threads here that I am seeking to interface my Osmocom-based GSM network to USA PSTN. In this post I shall provide details as to exactly what I am doing and how I go about it. I am currently experimenting with using bulkvs.com as my USA PSTN connectivity provider, chosen on the basis of cost economy: it is incredibly, mind-boggling cheap - the monthly "keepalive" cost is only 6 cents (yes, *cents*) per rented phone number! With phone numbers this cheap, I don't have to do any kind of "number economy" options: no need for "phone number NAT" (a single external phone number where a PBX answers, requiring callers to then dial an internal extension number), and no need for a mixed scheme where only one or two test SIMs get real outside numbers while all others get only fake internal numbers - instead I can give a real valid NANP phone number to every test SIM on my test/R&D GSM network. For context, my test GSM network is envisioned as serving two purposes: one is for development and testing of GSM handsets - real phone numbers are less important here - and the other purpose is demonstration to gathered audiences (American patriots and in-person freedom lovers who aren't afraid of some silly virus) for "oooh and ahhh" - and for this second purpose, having real phone numbers for every test phone will be a big boost. The interface to USA PSTN provided by bulkvs.com and other similar low-cost services is a SIP trunk, hence what I need to implement is a form of GSM to SIP gateway. But here is the part where I deviate from what appears to be the established canon in Osmocom-based community cellular networks: I don't want to run a heavyweight PBX like Asterisk or FreeSwitch, instead I want to run only a fairly thin, lightweight layer of extra sw between my OsmoMSC instance (MNCC interface) and the public-Internet SIP interface provided by bulkvs, with the needed thin sw layer put together by me, combining newly written code specific to my use case with bits of code borrowed from elsewhere. Here is the code I have written / put together so far: https://www.freecalypso.org/hg/themwi-system-sw/ The README file at the top level of the code repository explains what is there and how it works, in basic terms. One of the major pieces I will need to implement is an RTP bridge MGW that will transcode between GSM codecs (FR1, EFR, AMR) on the GSM side and G.711 (PCMU or PCMA) on the PSTN-via-SIP side. PSTN-via-SIP connectivity providers such as bulkvs don't support any GSM codecs (not one of them), instead they only support G.711 and G.729. I won't be doing G.729 in ThemWi (transcoding from one lossy vocoder to another seems like a terrible idea), hence I will run with G.711 on my PSTN-over-Internet interface and I will need to implement a transcoder between GSM codec family and G.711. I reason that what I need to implement is probably quite similar to the transcoding function in classic GSM E1 TRAUs, except that in the RTP bridge environment I also have to deal with the added complexities of packet loss, reordering and silence suppression, unpleasantries of the packet world that didn't exist in the olden days of TDM luxury. I don't know whether or not a transcoding RTP bridge MGW is already implemented somewhere in the bowels of either Asterisk or FreeSwitch - the size and complexity of both "major sw" choices is quite intimidating and off-putting. If someone who happens to know the answer gives me a pointer, it will certainly be very appreciated, but if not, then some time later I will undertake the big effort to dig in both projects and look for this transcoder. If the needed transcoder implementation already exists somewhere, then I will need to lift it out of whichever project and integrate it into themwi-system-sw - and if there is no existing published-source implementation, then I will be the first to produce one. M~

1 0

Using a mixture of authentication-required and unauthenticated IMSIs
by Mychaela Falconia 30 Jun '22

30 Jun '22

Hello Osmocom community, A question: is it possible to configure OsmoMSC+OsmoHLR such that authentication is required for those IMSIs for which key material records are present in HLR subscriber db, yet allow unauthenticated access for certain other IMSIs (specific ones, *not* accept-all) that are entered into HLR subscriber db without keys? My use case: I generally operate my network with SIMs of my own issue, and I wish to continue enforcing authentication (and using ciphering) for all Themyscira SIMs. However, I would like to play with some old phone models which cannot be found unlocked - only operator-locked units can be found on ebay - and some of those historical operator- locked phones come with SIM cards in them, issued by historical operators who have bit the dust ages ago. It would be really nice if I could add the IMSIs of those defunct-operator SIMs to my HLR subscriber db and allow them to connect - which would have to be unauthenticated, as I don't know their Ki or even their A38 algorithm version - but do so _without_ dropping the authentication requirement for my own Themyscira IMSIs. Is what I seek possible with current OsmoMSC and OsmoHLR? And if it isn't possible with current code, any ideas as to what would be the right way to implement such a feature? TIA, Mychaela

1 0

RFC: Reviving OsmoDevCall
by Harald Welte 28 Jun '22

28 Jun '22

[please follow-up-to openbsc(a)lists.osmocom.org so we don't cross-post all related mails] Dear Osmocom community, OsmoDevCall used to be rather successful for quite some time in recent years, but recently has been suffering quite a bit due to insufficient people volunteering to present. Big thanks to all who did! Interestingly, there's no shortage of ideas of topics at https://osmocom.org/projects/osmo-dev-con/wiki/OsmoDevCall - but then many of the potential speakers did not have the interest or time to follow-up. Most recently, the last few instances have not been taking place due to a lack of volunteers during my holidays. Last, but not least, while during COVID lockdown winter "friday night 8pm" was a good idea, this of course is more difficult during the summer, when people are more likely want to go out the weekend. So, to summarize, let me ask some questions: * would you be interested in OsmoDevCall continuing? * which day/time/timezone would you prefer ? * would you be able and willing to volunteer to give at talk within the next 3 or so months? Any other suggestions for or around OsmoDevCall are of course also welcome. Thanks in advance, Harald -- - Harald Welte <laforge(a)gnumonks.org> http://laforge.gnumonks.org/ ============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)

2 3

dazed and confused.
by Keith 27 Jun '22

27 Jun '22

Often I struggle to understand memory allocations and the "weird" things of C, so please point out to me what I'm missing here (I must be wrong, because you guys would not miss this, right? cleared_ci = (struct osmo_mgcpc_ep_ci){ .ep = ep, .mgcp_client_fi = ci->mgcp_client_fi, .got_port_info = ci->got_port_info, .rtp_info = ci->rtp_info, .occupied = true, /* .pending = true follows below */ .verb = verb, .notify = { .fi = notify, .success = event_success, .failure = event_failure, .data = notify_data, } }; osmo_strlcpy(cleared_ci.label, ci->label, sizeof(cleared_ci.label)); osmo_strlcpy(cleared_ci.mgcp_ci_str, ci->mgcp_ci_str, sizeof(cleared_ci.mgcp_ci_str)); *ci = cleared_ci; LOG_CI_VERB(ci, LOGL_DEBUG, "notify=%s\n", osmo_fsm_inst_name(ci->notify.fi)); #define LOG_CI_VERB(ci, level, fmt, args...) do { \ if (ci->verb_info.addr[0]) \ LOG_CI(ci, level, "%s %s:%u: " fmt, \ osmo_mgcp_verb_name(ci->verb), ci->verb_info.addr, ci->verb_info.port, \ ## args); \ else \ LOG_CI(ci, level, "%s: " fmt, \ osmo_mgcp_verb_name(ci->verb), \ ## args); \ } while(0) How is ci->verb_info not being using uninitialized here? Would that explain random crashes with this code? https://osmocom.org/issues/5572

2 2

Contract / Support
by jeff palmer 17 Jun '22

17 Jun '22

Hi, I am looking for contracted support on a osmocom based project, for a few days of configuration/setup assistance on a 2G / GPRS configuration. There are a lot of moving parts in the osmocom projects and versions of individual projects. Best to grab an expert on the subject than for us to dive in quite so deep just to work through some configurations. Please email me with contact details and will reach out immediately! Jeffery Palmer GSE

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

OpenBSC