[USRP-users] [srslte-users] MBMS issue

Justin Tallon justin.tallon at softwareradiosystems.com
Thu Jan 16 11:32:42 EST 2020


Hey Borja,

 I was not able to replicate the errors you saw using iperf .

Looking at your conf files again, I notice you
have interpolate_subframe_enabled = false among other incorrect
configurations outlined in the documentation. did you pass in different
values in the command line?

Regards,
Justin

[  3] local 239.255.1.1 port 5001 connected with 172.16.0.254 port 41368
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total
Datagrams
[  3]  0.0- 1.0 sec   940 KBytes  7.70 Mbits/sec   0.910 ms    0/  655 (0%)
[  3]  1.0- 2.0 sec  1.14 MBytes  9.55 Mbits/sec   0.944 ms    0/  812 (0%)
[  3]  2.0- 3.0 sec  1.26 MBytes  10.6 Mbits/sec   0.903 ms    0/  901 (0%)
[  3]  3.0- 4.0 sec  1.14 MBytes  9.54 Mbits/sec   0.908 ms    0/  811 (0%)
[  3]  4.0- 5.0 sec  1.26 MBytes  10.6 Mbits/sec   0.924 ms    0/  901 (0%)
[  3]  5.0- 6.0 sec  1.14 MBytes  9.54 Mbits/sec   0.905 ms    0/  811 (0%)
[  3]  6.0- 7.0 sec  1.19 MBytes  9.98 Mbits/sec   0.883 ms    0/  849 (0%)
[  3]  7.0- 8.0 sec  1.21 MBytes  10.1 Mbits/sec   0.902 ms    0/  863 (0%)
[  3]  8.0- 9.0 sec  1.14 MBytes  9.55 Mbits/sec   1.301 ms    0/  812 (0%)
[  3]  9.0-10.0 sec  1.26 MBytes  10.5 Mbits/sec   0.944 ms    0/  897 (0%)
[  3] 10.0-11.0 sec  1.14 MBytes  9.57 Mbits/sec   0.925 ms    0/  814 (0%)
[  3] 11.0-12.0 sec  1.26 MBytes  10.6 Mbits/sec   0.904 ms    0/  899 (0%)
[  3] 12.0-13.0 sec  1.14 MBytes  9.55 Mbits/sec   0.910 ms    0/  812 (0%)
[  3] 13.0-14.0 sec  1.25 MBytes  10.5 Mbits/sec   0.947 ms    0/  890 (0%)
[  3] 14.0-15.0 sec  1.15 MBytes  9.67 Mbits/sec   0.920 ms    0/  822 (0%)

____
Justin Tallon Ph.D.

Software Radio Systems (SRS)
http://www.softwareradiosystems.com

+353-86-067-0753 | justin.tallon at softwareradiosystems.com


On Thu, Jan 16, 2020 at 5:00 PM Borja Iñesta Hernández <
borieher at iteam.upv.es> wrote:

> Hi Justin,
>
> I'm Borja, Álvaro's coworker. Any updates about the MBMS issue?
>
> Anyway, we are trying to run MBMS with four computers. Computer 1 runs
> srsepc and srsmbms, computer 2 and 3 runs srsenb each one and computer 4
> runs srsue.
>
> Our problem here is that seems like the srsmbms has to run in the same
> computer as the srsenb in order to connect. We checked the source code and
> the connection is through a TUN device. Is it true that srsmbms can only be
> run in the same computer? This will change our architecture because we
> would need to run one srsmbms in computer 2 and 3 instead of one in
> computer 1.
>
> Thanks,
>
> Borja.
>
>
> El 14/1/20 a las 10:13, Álvaro Ibáñez escribió:
>
> Hi Justin,
>
> We have changed the CFO parameter by changing the parameter [rf]
> freq_offset to -4000 Hz in the ue.conf file. And now the cfo is more or
> less -1 kHz in the ue trace. Is this the correct way to change the CFO
> parameter?
>
> Attached the LOGs and the iperf results. The config files are still the
> same.
>
> We are going to test the multiples eNBs connected to one EPC in different
> machines and we will let you know the results.
>
> KR
>
> Alvaro
> El 14/1/20 a las 9:07, Justin Tallon escribió:
>
> Hey Alvaro,
>
> Looking at the iperf you sent me I see some errors which do not match the
> logs so ill try to reproduce that here.
>
> In the meantime, I noticed that you have a very high CFO (more than 5KHz)
> it will help to adjust that so that it is less than 1KHz.
>
> In order to attach multiple enbs to a remote EPC you will need to
>
>    - change the gtp_bind_addr in each eNodeB  to the addr of the
>    interface over which the enb will connect to the epc
>    - change the mme_add to the IP of the machine of the EPC
>    - change to enb_id in one of the enodebs config
>
> Let me know if you have success with this approach and ill have a look at
> the documentation.
>
> Thanks,
> Justin
> ____
> Justin Tallon Ph.D.
>
> Software Radio Systems (SRS)
> http://www.softwareradiosystems.com
>
> +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>
>
> On Mon, Jan 13, 2020 at 4:55 PM Álvaro Ibáñez <alibla at iteam.upv.es> wrote:
>
>> Hi Justin,
>>
>> We are currently able to connect eNB and EPC in different machines. We
>> could tell you how, it is just to change IP addresses in enb.conf and
>> epc.conf, but the documentation is not correct.
>>
>> Otherwise, we are quite interested in the question that i asked you
>> before about connecting more than one eNBs to one EPC without leaving the
>> MBMS issues.
>>
>> KR
>>
>> Alvaro
>> El 13/1/20 a las 15:11, Álvaro Ibáñez escribió:
>>
>> Hi Justin,
>>
>> We have performed new test with iperf 3M. Attached the LOGs and the iperf
>> results.
>>
>> , we have just a doubt. It's possible to connect more than one eNB to one
>> EPC???
>>
>> In the other hand, we are also trying to connect one eNB to one EPC in
>> different machines following the instructions given in the documentation
>> and using the default configuration -->
>> https://docs.srslte.com/en/latest/srsepc/source/2_epc_getstarted.html#enbs-and-srsepc-on-separate-machines
>>
>> And we can't initialize the EPC.
>>
>> Thanks in advance
>>
>> KR
>>
>> Alvaro
>> El 13/1/20 a las 14:17, Justin Tallon escribió:
>>
>> Hey Alvaro,
>>
>> I still see the throughput being maxed out as before.
>>
>> Can you try a lower resolution or possibly try a iperf at 3M to confirm
>> you can get that traffic through completely without error?
>>
>> Thanks,
>> Justin
>> ____
>> Justin Tallon Ph.D.
>>
>> Software Radio Systems (SRS)
>> http://www.softwareradiosystems.com
>>
>> +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>
>>
>> On Mon, Jan 13, 2020 at 12:36 PM Álvaro Ibáñez <alibla at iteam.upv.es>
>> wrote:
>>
>>> Hi Justin,
>>>
>>> We have tested with the big buck bunny video, with low quality and the
>>> quality has been better than before but in the playback of the video some
>>> errors occur, such as the freezing of the image for a certain time or some
>>> pixelated areas of the image.
>>>
>>> Attached the new LOGs. The config files are the same as before.
>>>
>>> KR
>>>
>>> Alvaro
>>> El 13/1/20 a las 11:07, Justin Tallon escribió:
>>>
>>> Hey Alvaro,
>>>
>>> I see some KOs in the logs which represent CRC failures but not too many.
>>>
>>> What I do notice that could be causing an issue is the fact that the
>>> bandwidth of your chosen eMBMS configuration is reaching capacity quite
>>> often which could cause you issues as there isn't sufficient bandwidth to
>>> properly stream your video.
>>>
>>> Maybe you could try a very low-resolution video as a test? if you can
>>> successfully send this video we will know this is causing issues.
>>>
>>> Regards,
>>> Justin
>>>
>>>
>>>
>>> ____
>>> Justin Tallon Ph.D.
>>>
>>> Software Radio Systems (SRS)
>>> http://www.softwareradiosystems.com
>>>
>>> +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>>
>>>
>>> On Mon, Jan 13, 2020 at 10:52 AM Álvaro Ibáñez <alibla at iteam.upv.es>
>>> wrHi Justin,
>>>
>>>> You're right, i did not realize this thing ( subframe 0 and 5 are not
>>>> used for MBMS). In fact, the error of PMCH appears when we interrupt the
>>>> simulation. But, when we are trying to send and receive a video, the
>>>> quality of the video in the UE side is quite poor.
>>>>
>>>> In the same simulation we have performed iperf test and the video
>>>> transmission, by using:
>>>>
>>>> >>ffmpeg -i $HOME/Documentos/IRT_short.mp4 -f mpegts
>>>> udp://sgi_mb@239.255.1.0:1234
>>>>
>>>> I changed the MCS of PMCH to 10, in order to test with more robust
>>>> modulation, but the quality still poor.
>>>>
>>>> Find attached the config files, LOGs and the iperf results.
>>>> El 13/1/20 a las 9:53, Justin Tallon escribió:
>>>>
>>>> Hey Alvaro,
>>>>
>>>> I think the reason for this is that there will never be eMBMS subframes
>>>> at 0 or 5 as the standard only allows for eMBMS at certain subframes, 0 and
>>>> 5 are never possible for eMBMS.
>>>>
>>>> Let me have a look into it anyway and send on logs and configs whenever
>>>> you can and ill have a look at them too!
>>>>
>>>> Thanks,
>>>> Justin
>>>>
>>>>
>>>> ____
>>>> Justin Tallon Ph.D.
>>>>
>>>> Software Radio Systems (SRS)
>>>> http://www.softwareradiosystems.com
>>>>
>>>> +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>>>
>>>>
>>>> On Mon, Jan 13, 2020 at 9:40 AM Álvaro Ibáñez Latorre <
>>>> alibla at iteam.upv.es> wrote:
>>>>
>>>>> Hi Justin,
>>>>>
>>>>> We forgot to save the LOGs and we were out of the office at the moment
>>>>> you answered.
>>>>>
>>>>> We are going to reproduce the error we faced and we will share you the
>>>>> LOGs and config files.
>>>>>
>>>>> Have you perform several PMCH test by using the pmch_test.c ??
>>>>>
>>>>> Thanks a lot
>>>>>
>>>>> Alvaro
>>>>>
>>>>>
>>>>>
>>>>> *Álvaro Ibáñez Latorre*
>>>>>
>>>>> Universitat Politècnica de València
>>>>>
>>>>> iTEAM Research Institute – Mobile Communications Group
>>>>>
>>>>> http://www.iteam.upv.es/?lang=es
>>>>>
>>>>>
>>>>>
>>>>> *De:* Justin Tallon <justin.tallon at softwareradiosystems.com>
>>>>> *Enviado el:* viernes, 10 de enero de 2020 14:49
>>>>> *Para:* Álvaro Ibáñez <alibla at iteam.upv.es>
>>>>> *CC:* Borja Iñesta Hernández <borieher at teleco.upv.es>
>>>>> *Asunto:* Re: [srslte-users] MBMS issue
>>>>>
>>>>>
>>>>>
>>>>> Hey Álvaro,
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Can you send on the config files including sib.conf you are using and
>>>>> also the resultant logs?
>>>>>
>>>>>
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Justin
>>>>>
>>>>> ____
>>>>> Justin Tallon Ph.D.
>>>>>
>>>>> Software Radio Systems (SRS)
>>>>> http://www.softwareradiosystems.com
>>>>>
>>>>> +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Fri, Jan 10, 2020 at 12:52 PM Álvaro Ibáñez <alibla at iteam.upv.es>
>>>>> wrote:
>>>>>
>>>>> Hi Justin.
>>>>>
>>>>> Thanks a lot for your answer and apologies for not answering you
>>>>> before.
>>>>>
>>>>> I saw the new release available two days ago and I have tested eMBMS
>>>>> with this release. Now I can run eMBMS but is it not stable and it seems to
>>>>> me that is not working correctly, i.e. sending and receiving video. I used
>>>>> the default configuration (config files, etc).
>>>>>
>>>>> In fact, when I tried to received a video, this error appeared int he
>>>>> UE side, it was related to PMCH channel, it's an error that I saw months
>>>>> ago. This is the error:
>>>>>
>>>>> When the number of symbols, for subframes 0 (control) or 5  is
>>>>> calculated in pmch.c function, this number is not equal to the number of
>>>>> symbols calculated in  ra_dl.c (I think). If you perform the pmch_test for
>>>>> subframe 0 you will observe the same as:
>>>>>
>>>>> I don't know if it is directly related to the instability.
>>>>>
>>>>> We have started a new project, called 5G ROSE, financied by ORCA and
>>>>> our patron is the TCD and we are going to perform network slicing, radio
>>>>> virtualization by using srsLTE and the Hypervisors available in the IRIS
>>>>> testbed of TCD.
>>>>>
>>>>> Thanks in advanced
>>>>>
>>>>> KR
>>>>>
>>>>> Alvaro
>>>>>
>>>>> El 14/11/19 a las 16:27, Justin Tallon escribió:
>>>>>
>>>>> Hey Alvaro,
>>>>>
>>>>>
>>>>>
>>>>> Apologies for the delay in solving your issue, I have been travelling
>>>>> for work.
>>>>>
>>>>> A temporary solution for you is to checkout this
>>>>> commit: c4bc151e14d88cfe30074a6a7fdc6591922c7abb.
>>>>>
>>>>>
>>>>>
>>>>> This is one before the commit which introduced this bug, this will
>>>>> allow you to continue to experiment while i fix and push a solution.
>>>>>
>>>>>
>>>>>
>>>>> Thanks for your patience!
>>>>>
>>>>>
>>>>>
>>>>> Justin
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ____
>>>>> Justin Tallon Ph.D.
>>>>>
>>>>> Software Radio Systems (SRS)
>>>>> http://www.softwareradiosystems.com
>>>>>
>>>>> +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Thu, Nov 7, 2019 at 1:13 PM Alvaro Ibañez Latorre <
>>>>> alibla at iteam.upv.es> wrote:
>>>>>
>>>>> Hi Justin,
>>>>> Thanks a lot for your answer.
>>>>> We will look forward for your updates.
>>>>> KR
>>>>> Alvaro
>>>>>
>>>>> Quoting Justin Tallon <justin.tallon at softwareradiosystems.com>:
>>>>>
>>>>> > Hey Alvaro,
>>>>> >
>>>>> > I know what the problem is now, thanks for drawing attention to this.
>>>>> >
>>>>> > I will get you a solution by the end of the week.
>>>>> >
>>>>> > Regards,
>>>>> > Justin
>>>>> >
>>>>> > ____
>>>>> > Justin Tallon Ph.D.
>>>>> >
>>>>> > Software Radio Systems (SRS)
>>>>> > http://www.softwareradiosystems.com
>>>>> >
>>>>> > +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>>>> >
>>>>> >
>>>>> > On Wed, Nov 6, 2019 at 5:18 PM Alvaro Ibañez Latorre <
>>>>> alibla at iteam.upv.es>
>>>>> > wrote:
>>>>> >
>>>>> >> Hi Justin
>>>>> >> Yes, I've attached both. See that we are running eNB with .example
>>>>> not
>>>>> >> with the normal enb.conf
>>>>> >> KR
>>>>> >> Alvaro
>>>>> >>
>>>>> >> Quoting Justin Tallon <justin.tallon at softwareradiosystems.com>:
>>>>> >>
>>>>> >> > Hey Alvaro,
>>>>> >> >
>>>>> >> > Can you please share the enb.conf file you have used and also the
>>>>> >> sib.conf?
>>>>> >> >
>>>>> >> > Thanks,
>>>>> >> >
>>>>> >> > Justin
>>>>> >> > ____
>>>>> >> > Justin Tallon Ph.D.
>>>>> >> >
>>>>> >> > Software Radio Systems (SRS)
>>>>> >> > http://www.softwareradiosystems.com
>>>>> >> >
>>>>> >> > +353-86-067-0753 | justin.tallon at softwareradiosystems.com
>>>>> >> >
>>>>> >> >
>>>>> >> > On Wed, Nov 6, 2019 at 4:37 PM Alvaro Ibañez Latorre <
>>>>> >> alibla at iteam.upv.es>
>>>>> >> > wrote:
>>>>> >> >
>>>>> >> >> Dear Community
>>>>> >> >> We are trying to run MBMS with the latest release available in
>>>>> the
>>>>> >> >> Github and we can't run the eNB when
>>>>> >> >> we are enabling MBMS. At first the manual is not updated to the
>>>>> latest
>>>>> >> >> release, so when we are trying to follow the manual instructions
>>>>> to
>>>>> >> >> enable MBMS is not possible, for example, expert in config file
>>>>> is not
>>>>> >> >> present, etc. Secondly, in order to run MBMS in the latest
>>>>> release, we
>>>>> >> >> are using the following instruction in the eNB:
>>>>> >> >> >> sudo srsenb enb.conf.example --scheduler.nof_ctrl_symbols 2
>>>>> >> >> >> --embms.enable true
>>>>> >> >> And it crashes after the eNodeB started message. We have
>>>>> attached an
>>>>> >> >> image and the backtrace.
>>>>> >> >> Thanks in advance.
>>>>> >> >> KR
>>>>> >> >> Alvaro
>>>>> >> >> _______________________________________________
>>>>> >> >> srslte-users mailing list
>>>>> >> >> srslte-users at lists.softwareradiosystems.com
>>>>> >> >>
>>>>> http://www.softwareradiosystems.com/mailman/listinfo/srslte-users
>>>>> >> >>
>>>>> >>
>>>>> >>
>>>>> >>
>>>>>
>>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20200116/a4df39b4/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ckacgpjdeogbnlog.png
Type: image/png
Size: 46575 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20200116/a4df39b4/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 8443 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20200116/a4df39b4/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.png
Type: image/png
Size: 86257 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20200116/a4df39b4/attachment-0002.png>


More information about the USRP-users mailing list