OpenBSC June 2009

openbsc@lists.osmocom.org

17 participants
103 discussions

patch 28: wrieshark pcap logging
by Andreas.Eversberg 10 Jun '09

10 Jun '09

This patch fixes the PCAP logging. If frames are received from E1 interface, an 8 byte mISDN header (MISDN_HEADER_LEN) is in front of the RSL packet. When frames are transmitted to the E1 interface, the pcap_write_packet function will get a message buffer without the mISDN header in front of the RSL packet. The pcap file was tested and the output is correct.

2 2

AW: layer 1 debugging
by Andreas.Eversberg 10 Jun '09

10 Jun '09

BN1E1 seems to have different layouts over the time. i always jumper the cards to TE, so i use a crosslink cable for NT to TE. ________________________________ Von: openbsc-bounces(a)lists.gnumonks.org [mailto:openbsc-bounces@lists.gnumonks.org] Im Auftrag von Eric Cathelinaud Gesendet: Mittwoch, 10. Juni 2009 10:41 An: Harald Welte Cc: openbsc(a)lists.gnumonks.org Betreff: Re: layer 1 debugging I tried to short the left and to invert the twinax plug (thus pluging them as in the wiki) and it works too. Then it could be a mistake in the jumper plan of BN1E1 cause they say that the left is "TE" and not "NT". Anyone else dealing with a BN1E1 card to confirm? Eric Cathelinaud 2009/6/9, Eric Cathelinaud <e.cathelinaud(a)googlemail.com>: 2009/6/9 Harald Welte <laforge(a)gnumonks.org> Hi Eric, On Tue, Jun 09, 2009 at 12:56:25PM +0200, Eric Cathelinaud wrote: > Thanks a lot for this layer 1 debugging tutorial. It helped me a lot since i > was thinking I did something wrong with mISDN. I checked the step 3 and find > out that my ISDN card was working fine. > Then I tried to invert my twinax plugs and it worked. I think there is a > mistake in the wiki. > It's written in http://bs11-abis.gnumonks.org/trac/wiki/BS11_Configuration : > - left Y01 PCM in (from BSC): RJ45 pin 4,5 > - Y02 PCM out (to BSC): RJ45 pin 1,2 > > but for me it works with : > Y01 PCM : RJ45 pin 1,2 (orange / white orange for a straight cable) > Y02 PCM : RJ45 pin 4,5 (blue / white blue for a straight cable) I think that depends on the jumper configuration of your E1 card. Did you really put the jumpers in "NT Mode" configuration? Yes I put NT Mode following the jumper plan (short pins on the right). My card is a Beronet BN1E1. I will try to short pins on the left and let you know if working as in wiki. > I can now see the network on my mobile phone thanks a lot. great! -- - 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

layer 1 debugging
by Andreas.Eversberg 10 Jun '09

10 Jun '09

hi, i added layer 1 state change (and initial state) reporting to mISDN kernel driver. this is usefull to resolve E1 connection problem. 1. download the latest mISDN driver from www.mISDN.org using git: git-clone git://git.misdn.org/git/mISDN.git/ change to socket branch: cd mISDN ./checkout-branch.sh socket compile and install as usual make force # remove kernel mISDN includes make make install load mISDN as usual. 2. run bsc_hack using debug option: --debug=DMI Allowing everyone? DB: Database initialized. DB: Database prepared. 1 devices found id: 0 Dprotocols: 00000018 Bprotocols: 0000006e protocol: 4 nrbchan: 30 name: hfc-e1.1 activate bchan activate bchan Mon Jun 1 18:51:43 2009 <1000> input/misdn.c:124 alen =6, dev(9) channel(0) sapi(63) tei(127) Mon Jun 1 18:51:43 2009 <1000> input/misdn.c:127 <= len = 12, prim(0x8) id(0x7f3f): DL_INFORMATION_IND Mon Jun 1 18:51:43 2009 <1000> input/misdn.c:134 DL_INFORMATION_IND: use channel(0) sapi(63) tei(127) for now mISDN message for unknown sign_link Mon Jun 1 18:51:43 2009 <1000> input/misdn.c:124 alen =6, dev(9) channel(0) sapi(63) tei(127) Mon Jun 1 18:51:43 2009 <1000> input/misdn.c:127 <= len = 8, prim(0x102) id(0x7f3f): PH_ACTIVATE_IND Mon Jun 1 18:51:43 2009 <1000> input/misdn.c:161 PH_ACTIVATE_IND: channel(0) sapi(63) tei(127) in this case we see "PH_ACTIVATE_IND", because BS11 is connected and sending a signal. 3. check your card: (most common type) close to the plug there are two jumpers. the are used to select between 1,2,4,5 and 1,2,3,6 pin selection. be sure they are to the left for standard 1,2,4,5 pinning. take an rj45 plug and connect pin 1 with 4 and 2 with 5. start bsc_hack and plug it into your card. you should see "PH_ACTIVATE_IND" when you plug it. your card is working. if nothing works, check your termination switches. refer to the card's manual. the must be set to 120 ohms, if the termination jumpers on your BS11 are pulled out. without termination, the card might not work. 4. check your BS11 remove the plug and connect the cable to the BS11: one pair (pin 1 and 2) to Y01 and one pair (pin 4 and 5) to Y02. after bootup ob BS11 you will see "PH_ACTIVATE_IND" if you connect both paris in the correct order. you must connect both pairs because BS11 will only send a signal, if it receives a signal. pin 1 can be wapped with pin 2, pin 4 with pin 5 also. the E1 interface does not care about that. 5. check data link connection: you have already received the management process via "DL_INFORMATION_IND" on sapi 63 tei 127. if BS11 software load is running, you will receive additional data links from BS11: - sapi 62 tei 25 - sapi 0 tei 25 - sapi 0 tei 1 these are three datalinks. each one is created and sends us "DL_INFORMATION_IND" on creation, "DL_ESTABLISH_IND" on activation and "DL_DATA_IND" when receiving data on that link. hope that helps. andreas

3 5

patch 35: information element transcoding
by Andreas.Eversberg 10 Jun '09

10 Jun '09

this patch will do the transcoding of information elements. this patch adds new functions to be used for the application interface patch, and is basis for the application patch (currently patch 36). when applying the patch, the compiler gives warnings, because the "static" functions will not be used. (yet)

2 1

patch 33: db fix
by Andreas.Eversberg 09 Jun '09

09 Jun '09

during attachment, the "tmsi" field of the subscriber may be an empty string, so the string must be quoted in the sql request, or the sql request fails due to incorrect syntax. also i added "extension" field to debug output.

1 0

inofficioal test
by Andreas.Eversberg 09 Jun '09

09 Jun '09

hi, i uploaded an inofficial test version of openbsc with all my patches applied. (not the latest, but working well). you will find it here: http://www.linux-call-router.de/download/lcr-1.5/ <http://www.linux-call-router.de/download/lcr-1.5/> all ressources get freed, no hanging logical channels or subscriber instances in any test case i found. also linux-call-router, an alternative application to bsc_hack, the gsm speech codec and mISDN kernel drivers are available there. a howto is in progress. regards, andreas

2 1

patch 31: fixed option for db selection
by Andreas.Eversberg 09 Jun '09

09 Jun '09

the database select option "-l" did not work: - c = getopt_long(argc, argv, "hc:n:d:sar:p:f:t:C:RL:", + c = getopt_long(argc, argv, "hc:n:d:sar:p:f:t:C:RL:l:",

1 0

patch 37: assigning traffic channel
by Andreas.Eversberg 09 Jun '09

09 Jun '09

hi, i have a phone that requests a SDCCH channel after it is paged. (even if paging request required a TCH Full channel.) in this case, i assign a GSM_LCHAN_TCF_F channel except for location update. (see patch) in the second part of this patch, i check if the current channel is a traffic channel. if not, we can't use it on calls BSS->MS. this happens if a call is received after location update, but before the channel is released. this second part depends on the application patch (36). andreas

2 1

patch: 30
by Andreas.Eversberg 09 Jun '09

09 Jun '09

my subsequent patches are dependant on the patches by holger from may 5th. for personal testing and for information, i packed them together. i tested them and they work well. Mit freundlichen Grüßen, .-. /'v'\ (/ \) ------------------------------------------------------------------"-"- |_| i.A. Andreas Eversberg Administrator Betrieb und Entstörung Versatel Nord GmbH Nordstr. 2 D-24937 Flensburg Fon: +49-461-9099749 | Fax: +49-461-909960749 andreas.eversberg@versatel.de@versatel.de | www.versatel.de Sitz der Gesellschaft: Flensburg, Registergericht: Flensburg, HRB 3395 FL Geschäftsführer: Soeren Wendler, Dr. Hai Cheng, Dr. Max Padberg, Peter Schindler, Joachim Bellinghoven

2 1

patch 32: subscr_update() missing.
by Andreas.Eversberg 09 Jun '09

09 Jun '09

If location update is requested, but subscriber is not yet authorised within mm_rx_loc_upd_req() function, the subscr_update() is not called, because subscriber information is not complete. During mm_rx_id_resp() the subscriber informations is may be complete, so authorize subscriber succeeds and database must be updated.

2 1

← Newer
1
...
4
5
6
7
8
9
10
11
Older →

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

OpenBSC June 2009