Dear Osmocom list
We are getting several audio problems and dropped calls. The audio sounds like static. We are also getting large frame correction in the logs, and SACCH deactivation timeout.
Any ideas what could be the likely culprit?
2017-03-24_14:30:57.93546 <001a> rtp_proxy.c:295 Correcting frame difference of 74518293571 frames 2017-03-24_14:31:11.45899 <001a> rtp_proxy.c:295 Correcting frame difference of 1 frames 2017-03-24_14:31:11.51590 <001a> rtp_proxy.c:295 Correcting frame difference of 2 frames 2017-03-24_14:31:11.65513 <001a> rtp_proxy.c:295 Correcting frame difference of 1 frames
2017-03-24_14:32:08.95028 <0004> abis_rsl.c:1343 (bts=1,trx=0,ts=1,ss=0) SACCH deactivation timeout. 2017-03-24_14:37:10.11844 <0004> abis_rsl.c:662 (bts=1,trx=0,ts=1,ss=0) is back in operation.
On 24 Mar 2017, at 15:49, Edwin Reed-Sanchez ereedsanchez@gmail.com wrote:
Dear Osmocom list
We are getting several audio problems and dropped calls. The audio sounds like static. We are also getting large frame correction in the logs, and SACCH deactivation timeout.
Any ideas what could be the likely culprit?
classic radio interference?
2017-03-24_14:30:57.93546 <001a> rtp_proxy.c:295 Correcting frame difference of 74518293571 frames
looks like a change of SSRC? What does RTP look like?
2017-03-24_14:31:11.45899 <001a> rtp_proxy.c:295 Correcting frame difference of 1 frames 2017-03-24_14:31:11.51590 <001a> rtp_proxy.c:295 Correcting frame difference of 2 frames 2017-03-24_14:31:11.65513 <001a> rtp_proxy.c:295 Correcting frame difference of 1 frames
2017-03-24_14:32:08.95028 <0004> abis_rsl.c:1343 (bts=1,trx=0,ts=1,ss=0) SACCH deactivation timeout. 2017-03-24_14:37:10.11844 <0004> abis_rsl.c:662 (bts=1,trx=0,ts=1,ss=0) is back in operation.
Holger,
classic radio interference?
- could this be caused by having amplifiers to high, and being close to another BTS? Should I attenuate the amplification in Osmotrx?
looks like a change of SSRC? What does RTP look like?
- where do I gather information for RTP. Is it in the logs, or in the configs. We currently are not using rtp jitter-buffer 0 in the osmobts config.
We are also getting radio link failures.
e2017-03-24_14:58:33.56803 <0004> abis_rsl.c:1035 (bts=1,trx=0,ts=1,ss=0) CONNECTION FAIL: RELEASING state ACTIVE CAUSE=0x01(Radio Link Failure).
Will send you files in separate email.
Thanks
On Fri, Mar 24, 2017 at 10:52 AM, Holger Freyther holger@freyther.de wrote:
On 24 Mar 2017, at 15:49, Edwin Reed-Sanchez ereedsanchez@gmail.com
wrote:
Dear Osmocom list
We are getting several audio problems and dropped calls. The audio sounds like static. We are also getting large frame
correction in the logs, and SACCH deactivation timeout.
Any ideas what could be the likely culprit?
classic radio interference?
2017-03-24_14:30:57.93546 <001a> rtp_proxy.c:295 Correcting frame
difference of 74518293571 frames
looks like a change of SSRC? What does RTP look like?
2017-03-24_14:31:11.45899 <001a> rtp_proxy.c:295 Correcting frame
difference of 1 frames
2017-03-24_14:31:11.51590 <001a> rtp_proxy.c:295 Correcting frame
difference of 2 frames
2017-03-24_14:31:11.65513 <001a> rtp_proxy.c:295 Correcting frame
difference of 1 frames
2017-03-24_14:32:08.95028 <0004> abis_rsl.c:1343 (bts=1,trx=0,ts=1,ss=0)
SACCH deactivation timeout.
2017-03-24_14:37:10.11844 <0004> abis_rsl.c:662 (bts=1,trx=0,ts=1,ss=0)
is back in operation.
On 24 Mar 2017, at 16:09, Edwin Reed-Sanchez ereedsanchez@gmail.com wrote:
Holger,
classic radio interference?
- could this be caused by having amplifiers to high, and being close to another BTS? Should I attenuate the amplification in Osmotrx?
Difficult to say. You can look at the RSL measurement reports and take a look if there is something obvious, e.g. rxlev/rxqual being very bad
looks like a change of SSRC? What does RTP look like?
- where do I gather information for RTP. Is it in the logs, or in the configs. We currently are not using
rtp jitter-buffer 0 in the osmobts config.
In Wireshark under Edit->Preferences->Protocol->RSL ("Use nanoBTS defitnition"). Then you will see IPA CRCX/MDCX/DLCX to manage the audio flow and Wireshark will be able to identify the RTP streams.
With Telephony->RTP->Show all Streams you can see how many streams SRC ip/port, DST ip/port, SSRC exist. Then compare the timestamps from the log with timestamp of the RTP packet. Is there another stream/SSRC? Is this is just the first sequence number (which should be random) and leads to a big jump?
holger
-
Edwin Reed-Sanchez -
Holger Freyther