OpenBSC July 2010

openbsc@lists.osmocom.org

19 participants
48 discussions

Detect phone calls
by Luca Bertoncello 26 Jul '10

26 Jul '10

Hi, list! Of course OpenBSC knows which phone-calls there are right now. I'd like to get this list. How can I do this? Thanks a lot! -- _______________________________________________________________________ Luca Bertoncello Entwicklung Mail: bertoncello(a)netzing.de NETZING Solutions AG Tel.: 0351/41381 - 23 Kesselsdorfer Str. 216, 01169 Dresden Fax: 0351/41381 - 12 HRB 18926 / Ust.ID DE211326547 Mail: netzing.ag(a)netzing.de _______________________________________________________________________

3 4

Re: PATCH/RFC use recvfrom with MSG_TRUNC as flag
by Dieter Spaar 24 Jul '10

24 Jul '10

Hello Holger, On Fri, 23 Jul 2010 21:38:36 +0800, "Holger Hans Peter Freyther" <holger(a)freyther.de> wrote: > > PS: How big is your diff to make it compile on Windows? I have not ported the TUN calls yet. My last try for OpenBSC plus GPRS was in a Linux VM, however I had problems there with the TUN interface. I could see outgoing packets with proper NAT translation (the DNS queries from the phone) and I could also see the DNS replies "outside" the VM but they never got through to the TUN interface. Maybe its a problem related to the VM, but after this experience I decided to wait playing with the TUN interface on Cygwin because it probably makes sense to have a running reference system first. BTW, anyone having OpenBSC plus GPRS running on MacOS X ? Thats the closest to Unix I have here besides a VM. Best regards, Dieter -- Dieter Spaar, Germany spaar(a)mirider.augusta.de

1 0

[PATCH] Subscriber data in measurement report
by Luca Bertoncello 24 Jul '10

24 Jul '10

I find the measurement report of OpenBSC very difficult to read, because it is not possible to distinguish the various phones. Of course, it is possible to filter the log (logging filter imsi ...), but sometimes, as in my case, it is necessary to get the measurement reports of many phones at the same time. In this case, it is not possible to know which phone generate which report. I create a patch to solve this problem. With this patch, the measurement reports contain the lchan in which the call (or SMS) runs (just as information), the IMSI and the extension of the phone. If you are interested, here is it! Greetings -- _______________________________________________________________________ Luca Bertoncello Entwicklung Mail: bertoncello(a)netzing.de NETZING Solutions AG Tel.: 0351/41381 - 23 Fröbelstr. 57, 01159 Dresden Fax: 0351/41381 - 12 HRB 18926 / Ust.ID DE211326547 Mail: netzing.ag(a)netzing.de _______________________________________________________________________

1 0

Re: PATCH/RFC use recvfrom with MSG_TRUNC as flag
by Dieter Spaar 23 Jul '10

23 Jul '10

Hello Holger, On Fri, 23 Jul 2010 19:07:55 +0800, "Holger Hans Peter Freyther" <holger(a)freyther.de> wrote: > > Dieter do you know if cygwin will silently ignore the MSG_TRUNC > flag or if you will have a problem on Window? Thanks for asking. According to what I found so far, Cygwin should support MSG_TRUNC. However considering the fact that there is much more to do to get OpenBSC and GPRS running under Cygwin (porting the calls to access the TUN interface in OpenGGSN), one more obstacle should not really care ;-) (I guess I am the only one who uses OpenBSC on Cygwin). Best regards, Dieter -- Dieter Spaar, Germany spaar(a)mirider.augusta.de

2 1

PATCH/RFC use recvfrom with MSG_TRUNC as flag
by Holger Hans Peter Freyther 23 Jul '10

23 Jul '10

Hi Harald, Dieter, what do you think of using MSG_TRUNC and checking if we have truncated our packets in a recv call? Dieter do you know if cygwin will silently ignore the MSG_TRUNC flag or if you will have a problem on Window? diff --git a/openbsc/src/gprs/gprs_ns.c b/openbsc/src/gprs/gprs_ns.c index 3db1d67..ab2d937 100644 --- a/openbsc/src/gprs/gprs_ns.c +++ b/openbsc/src/gprs/gprs_ns.c @@ -867,7 +867,7 @@ static struct msgb *read_nsip_msg(struct bsc_fd *bfd, int *error, return NULL; } - ret = recvfrom(bfd->fd, msg->data, NS_ALLOC_SIZE - NS_ALLOC_HEADROOM, 0, + ret = recvfrom(bfd->fd, msg->data, NS_ALLOC_SIZE - NS_ALLOC_HEADROOM, MSG_TRUNC, (struct sockaddr *)saddr, &saddr_len); if (ret < 0) { LOGP(DNS, LOGL_ERROR, "recv error %s during NSIP recv\n", @@ -879,6 +879,11 @@ static struct msgb *read_nsip_msg(struct bsc_fd *bfd, int *error, msgb_free(msg); *error = ret; return NULL; + } else if (ret > NS_ALLOC_SIZE - NS_ALLOC_HEADROOM) { + LOGP(DNS, LOGL_ERROR, "truncated msg. real size is: %d\n", ret); + msgb_free(msg); + *error = -ENOSPC; + return NULL; } msg->l2h = msg->data;

1 0

Problem to compile the last version of OpenBSC
by Luca Bertoncello 23 Jul '10

23 Jul '10

Hi, list! A very strange problem... Today I got a new BTS (DSC1800) and I tried to configure it. Unfortunately it does not respond correctly (I'll send a separate E-Mail about this problem), so I decided to try with the last version of OpenBSC (maybe a bug in ipaccess-config?). Unfortunately I can't compile it. I get these errors: gcc -Wall -I/usr/local/include/ -I/usr/local/include/ -g -O2 -L/usr/local/lib -losmocore -o bsc_hack bsc_hack.o bsc_init.o bsc_vty.o vty_interface_layer3.o libmsc.a libbsc.a libvty.a libmsc.a -ldl -ldbi -L/usr/local/lib -losmovty -lcrypt bsc_init.o:(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here bsc_vty.o:(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here vty_interface_layer3.o:(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(gsm_subscriber.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(db.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(mncc.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(gsm_04_08.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(gsm_04_11.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(transaction.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(token_auth.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(rrlp.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(ussd.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(silent_call.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(handover_decision.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(auth.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(osmo_msc.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libmsc.a(gsm_04_80.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(abis_rsl.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(abis_nm.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(gsm_data.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(gsm_04_08_utils.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(chan_alloc.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(debug.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(abis_nm_vty.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(gsm_subscriber_base.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(bsc_rll.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(trau_mux.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(paging.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(e1_config.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(e1_input.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(misdn.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(ipaccess.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(handover_logic.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(talloc_ctx.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(system_information.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(rest_octets.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(rtp_proxy.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(bts_siemens_bs11.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(bts_ipaccess_nanobts.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(bts_unknown.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(bsc_api.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(meas_rep.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(socket.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libbsc.a(subchan_demux.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here libvty.a(common_vty.o):(.data+0x0): multiple definition of `gsm_7bit_alphabet' bsc_hack.o:(.data+0x0): first defined here collect2: ld gab 1 als Ende-Status zurück make[3]: *** [bsc_hack] Fehler 1 make[3]: Verlasse Verzeichnis '/home/lucabert/BSC/openbsc/openbsc/src' make[2]: *** [all-recursive] Fehler 1 make[2]: Verlasse Verzeichnis '/home/lucabert/BSC/openbsc/openbsc/src' make[1]: *** [all-recursive] Fehler 1 make[1]: Verlasse Verzeichnis '/home/lucabert/BSC/openbsc/openbsc' make: *** [all] Fehler 2 Can someone confirm me, that the problem is in this version and not in my PC? Thanks -- _______________________________________________________________________ Luca Bertoncello Entwicklung Mail: bertoncello(a)netzing.de NETZING Solutions AG Tel.: 0351/41381 - 23 Fröbelstr. 57, 01159 Dresden Fax: 0351/41381 - 12 HRB 18926 / Ust.ID DE211326547 Mail: netzing.ag(a)netzing.de _______________________________________________________________________

2 2

libosmocore now abort()s on msgb overflow/underflow
by Harald Welte 23 Jul '10

23 Jul '10

Hi! I've just committed a patch that will cause libosmocore to abort in case we try to msgb_put() beyond the end fo the buffer or msgb_push() ahead of the start. I hope we can uncover any hidden buffer under/overflows this way, if they exist. Regards, 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)

1 0

installing openbsc on opensuse
by jason mercury 23 Jul '10

23 Jul '10

Hi to all, I want to use openbsc on opensuse linux. What procedure shoul i follow? Thanks. Jason

2 1

Switch with PowerOverEthernet for a BaseStation
by Luca Bertoncello 21 Jul '10

21 Jul '10

Hi, list! We have actually 4 BaseStation (but in the future we will buy at least other 2). Now want my Boss to buy a PoE-Switch at which we connect all BaseStation. This Switch has to supply power, too. Now, I'm not sure (I don't know PoE in detail) if I can use every PoE-Switch. I searched in Internet and I found a Netgear FS108P. It's cheap and it will be enough for us. Can I use it with our BaseStation or have I to use the power supplier of the BaseStations, too? Thanks a lot for your answers! -- _______________________________________________________________________ Luca Bertoncello Entwicklung Mail: bertoncello(a)netzing.de NETZING Solutions AG Tel.: 0351/41381 - 0 Kesselsdorfer Str. 216, 01169 Dresden Fax: 0351/41381 - 12 HRB 18926 / Ust.ID DE211326547 Mail: netzing.ag(a)netzing.de _______________________________________________________________________

4 6

[PATCH] tests: don't hardcode length values of expected encoding gsm_7bit_encode: make sure to return the number of actually written bytes gsm_7bit_decode: calculate length of resulting septets from input length before decoding
by Nico Golde 20 Jul '10

20 Jul '10

The input length to gsm_7bit_decode reflects the number of encoded bytes to be decoded. As the decoding is done on the input in septetes we need to take this into account and recalculate the length. --- src/gsm_utils.c | 10 ++++++---- tests/sms/sms_test.c | 21 ++++++++++----------- 2 files changed, 16 insertions(+), 15 deletions(-) diff --git a/src/gsm_utils.c b/src/gsm_utils.c index 3a378ac..818655e 100644 --- a/src/gsm_utils.c +++ b/src/gsm_utils.c @@ -49,21 +49,23 @@ static int gsm_septet_lookup(uint8_t ch) /* GSM 03.38 6.2.1 Character unpacking */ int gsm_7bit_decode(char *text, const uint8_t *user_data, uint8_t length) { - int i = 0; + int i = 0,y; int l = 0; + int septet_l = (length * 8) / 7; uint8_t *rtext = calloc(length, sizeof(uint8_t)); uint8_t tmp; /* FIXME: We need to account for user data headers here */ i += l; - for (; i < length; i ++){ + for (; i < septet_l; i++){ rtext[i] = ((user_data[(i * 7 + 7) >> 3] << (7 - ((i * 7 + 7) & 7))) | (user_data[(i * 7) >> 3] >> ((i * 7) & 7))) & 0x7f; } - for(i = 0; i < length; i++){ + + for(i = 0; i < septet_l; i++){ /* this is an extension character */ if(rtext[i] == 0x1b && i + 1 < length){ tmp = rtext[i+1]; @@ -139,7 +141,7 @@ int gsm_7bit_encode(uint8_t *result, const char *data) } free(rdata); - return i; + return z; } /* determine power control level for given dBm value, as indicated diff --git a/tests/sms/sms_test.c b/tests/sms/sms_test.c index 3742dd8..9d87b5b 100644 --- a/tests/sms/sms_test.c +++ b/tests/sms/sms_test.c @@ -37,7 +37,7 @@ struct test_case { static const char simple_text[] = "test text"; static const uint8_t simple_enc[] = { - 0xf4, 0xf2, 0x9c, 0x0e, 0xa2, 0x97, 0xf1, 0x74, 0x00, + 0xf4, 0xf2, 0x9c, 0x0e, 0xa2, 0x97, 0xf1, 0x74 }; static const char escape_text[] = "!$ a more#^- complicated test@@?_\%! case"; @@ -46,7 +46,6 @@ static const uint8_t escape_enc[] = { 0x86, 0xd2, 0x02, 0x8d, 0xdf, 0x6d, 0x38, 0x3b, 0x3d, 0x0e, 0xd3, 0xcb, 0x64, 0x10, 0xbd, 0x3c, 0xa7, 0x03, 0x00, 0xbf, 0x48, 0x29, 0x04, 0x1a, 0x87, 0xe7, 0x65, - 0x00, 0x00, 0x00, 0x00, 0x00 }; static const struct test_case test_encode[] = @@ -54,12 +53,12 @@ static const struct test_case test_encode[] = { .input = simple_text, .expected = simple_enc, - .expected_length = 9, + .expected_length = sizeof(simple_enc), }, { .input = escape_text, .expected = escape_enc, - .expected_length = 41, + .expected_length = sizeof(escape_enc), }, }; @@ -67,12 +66,12 @@ static const struct test_case test_decode[] = { { .input = simple_enc, - .input_length = 9, + .input_length = sizeof(simple_enc), .expected = simple_text, }, { .input = escape_enc, - .input_length = 41, + .input_length = sizeof(escape_enc), .expected = escape_text, }, }; @@ -89,13 +88,13 @@ int main(int argc, char** argv) char result[256]; /* test 7-bit encoding */ - for (i = 0; i < ARRAY_SIZE(test_encode); ++i) { - memset(coded, 0, sizeof(coded)); + for (i = 0; i < ARRAY_SIZE(test_encode); ++i) { + memset(coded, 0x42, sizeof(coded)); length = gsm_7bit_encode(coded, test_encode[i].input); if (length != test_encode[i].expected_length) { - fprintf(stderr, "Failed to encode case %d. Got %d\n", - i, length); + fprintf(stderr, "Failed to encode case %d. Got %d, expected %d\n", + i, length, test_encode[i].expected_length); return -1; } @@ -108,7 +107,7 @@ int main(int argc, char** argv) /* test 7-bit decoding */ for (i = 0; i < ARRAY_SIZE(test_decode); ++i) { - memset(result, 0, sizeof(coded)); + memset(result, 0x42, sizeof(coded)); gsm_7bit_decode(result, test_decode[i].input, test_decode[i].input_length); -- 1.7.1 --7JfCtLOvnd9MIVvH--

1 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

OpenBSC July 2010