Further clarification: Regarding my initial mixup of standard iub / abis, it was because of this, when i first met the problem i googled for a solution and ran into this:
https://osqa-ask.wireshark.org/questions/23531/abis-over-iua-decoding it was actually abis over IuA. On Sat, Jun 2, 2018, 14:57 Borjan Jovanovski <jovanovskibor...@gmail.com> wrote: > Hi Harald, > > We are talking about a ZTE MacroBTS ZXSDR BS8900A. > > And about your further questions, your final statement is correct it is > standard abis dialect encapsulated in SCTP, i can provide documentation at > a later time as i am not home at the moment, but i can tell you now that, > the SCTP protocol is terminated in a switch type device before the bsc > hardware, and then further communication continues over tcp to the bsc. > Regards, > Borjan > > On Sat, Jun 2, 2018, 14:50 Harald Welte <lafo...@gnumonks.org> wrote: > >> Hi Borjan, >> >> On Sat, Jun 02, 2018 at 02:15:38PM +0200, Borjan Jovanovski wrote: >> >> > my device is a multimode device BTS NodeB and eNodeB 2G + 3G + 4G and >> > uses SCTP protocol for BSC RNC and MME connection >> >> Can you tell us more about that device? >> >> > ... but i couldn't get OpenBSC to work with abis over sctp. So my final >> question is: >> > How to handle Abis over SCTP with OpenBSC. >> >> To the best of my knowledge, there is no 3GPP specification for Abis over >> SCTP. >> >> As such, it's not a surprise that we're not implementing it in OsmoBSC :) >> Maybe I missed something. Can you please point me to relevant >> specification for this Abis dialect? >> >> I would assume it's some vendor-specific dialect. Do you have >> protocol-level >> documentation for that available? Do you have protocol traces of this >> dialect? >> >> In any case, even with documentation and/or traces, you would need to >> develop the respective code to interface this new Abis >> dialect/encapsulation. The required effort depends on how far that >> dialect diverges from the existing "known" dialects/formats. If it's >> just 08.58 RSL and 12.21 OML inside SCTP, then it should be rather >> little effort. But that's a big "if" :) >> >> Regards, >> Harald >> -- >> - Harald Welte <lafo...@gnumonks.org> >> http://laforge.gnumonks.org/ >> >> ============================================================================ >> "Privacy in residential applications is a desirable marketing option." >> (ETSI EN 300 175-7 Ch. >> A6) >> >
