Toggle sidebar Toggle sidebar

3cx audio drops

Status
Not open for further replies.

Cody Fontaine

Joined
Jun 16, 2017
Messages
43
Reaction score
0
My set-up:
Comcast Cisco DPC3941B Business Gateway
cisco rv220w router

The cisco rv220w is in DMZ on Comcast Gateway
the SBC is in DMZ on Rv220w

No SIP ALG enabled on either as far as I can tell.

30 extensions
Fanvil x3 x4 x5
yealink 21p_e2
Windows Softphones


I have a 3cx SBC on premise. It is a
intel core i3 550 3.2 ghz dual core
4gb ram
debian 9

The server is a google compute engine
g1-small 1 vcpu 1.7gb ram intel ivy bridge


My problem: Audio drops out so that the caller cannot here the callee inbound and outbound.
internal and external calls

errors I see in the log
Code: 
[CM505001]: Endpoint Extn:307: Device info: Device Not Identified: User Agent not matched; Capabilities:[reinvite, replaces, able-no-sdp, recvonly] UserAgent: [3CXPhone for Windows 15.5.12227.5] PBX contact: [sip:[email protected]:5060]

10:52:32.225|7f298cd1d700|Error|/home/repomaster/workspace/15.5SP6/SPBuild/Sources/Projects/Tunnel/SipCore/SipCore.cpp(708): OK w/out SDP: SipResp: 20

11:02:01.250|7f45934a7700| Warn|/home/repomaster/workspace/15.5SP6/SPBuild/Sources/Projects/MediaServer/src/RTPReceiver.cpp(440): 5:MediaProxy:VRTP receiver for [email protected]:5482(C:49.2) has no assigned socket
11:02:01.250|7f45934a7700| Warn|/home/repomaster/workspace/15.5SP6/SPBuild/Sources/Projects/MediaServer/src/RTPReceiver.cpp(440): 5:MediaProxy:VRTCP receiver for [email protected]:5482(C:49.2) has no assigned socket
11:02:01.250|7f45934a7700| Warn|/home/repomaster/workspace/15.5SP6/SPBuild/Sources/Projects/MediaServer/src/RTPReceiver.cpp(440): 5:MediaProxy:T38 receiver for [email protected]:5482(C:49.2) has no assigned socket
11:02:01.250|7f45934a7700| Warn|/home/repomaster/workspace/15.5SP6/SPBuild/Sources/Projects/MediaServer/src/RTPReceiver.cpp(440): 5:MediaProxy:T38_ receiver for [email protected]:5482(C:49.2) has no assigned socket

I have attached a wireshark as well

The wireshark does not show any RTP streams

all firewall tests clear
no sip alg

Most calls internal and external drop audio between 5 and 10 minutes. However some calls can go for hours without dropping inconsistently. Across all devices
 

Attachments

Last edited:
I myself seem to be experiencing a similar issue that really seems to have just started over the last 3 weeks we have a n onprem 3cx solution. I have seem to have narrowed some stuff down to having to restart to 3CX Media Server services every day because of the system dropping incoming and outgoing audio. We are no leveraging SIP ALG either. Using straight NAT through the firewall. Solution has been working for 1.5 years but has recently become very flaky.

here's some of the output from our 3CXMediaServer Logs:
Code: 
11:59:54.599|00004690| Info|MSEndPoint.cpp(2606): 5:[MS205001] C:509.1 RTP: Party address changed from 67.231.1.79:23372 to 67.231.1.79:1037
11:59:54.634|000022fc|Trace|MSEndPoint.cpp(2633): 5:[MS505003] C:509.4: RTP stream from 10.1.1.214:12164 restored.
11:59:54.660|00003c30|Trace|MSCallConf.cpp(165): 0:Call: [email protected]:5482 destroyed
12:00:00.754|00003ec4|Trace|MediaServer.cpp(1687): 4:EndPoint [email protected]:5482 removed from call [email protected]:5482
12:00:00.755|00003ec4|Trace|MediaServer.cpp(1416): 9:references to EndPoint [email protected]:5482 were removed
12:00:00.755|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:VRTP receiver for [email protected]:5482(C:509.1) has no assigned socket
12:00:00.755|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:VRTCP receiver for [email protected]:5482(C:509.1) has no assigned socket
12:00:00.755|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:T38 receiver for [email protected]:5482(C:509.1) has no assigned socket
12:00:00.755|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:T38_ receiver for [email protected]:5482(C:509.1) has no assigned socket
12:00:00.757|00003ec4|Trace|MediaServer.cpp(1687): 4:EndPoint [email protected]:5482 removed from call [email protected]:5482
12:00:00.758|00003ec4|Trace|MediaServer.cpp(1416): 9:references to EndPoint [email protected]:5482 were removed
12:00:00.758|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:VRTP receiver for [email protected]:5482(C:509.4) has no assigned socket
12:00:00.758|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:VRTCP receiver for [email protected]:5482(C:509.4) has no assigned socket
12:00:00.758|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:T38 receiver for [email protected]:5482(C:509.4) has no assigned socket
12:00:00.758|00004640| Warn|RTPReceiver.cpp(440): 5:MediaProxy:T38_ receiver for [email protected]:5482(C:509.4) has no assigned socket
12:00:00.762|000022fc|Trace|RTPReceiver.cpp(167): 5:RTPReceiver[1]:RTP receiver was removed for [email protected]:5482(C:509.4) on socket (652)
12:00:00.762|000022fc|Trace|RTPReceiver.cpp(167): 5:RTPReceiver[1]:RTCP receiver was removed for [email protected]:5482(C:509.4) on socket (656)
12:00:00.774|00003ae0|Trace|MSEndPoint.cpp(818): 0:EndPoint [email protected]:5482 destroyed
12:00:00.774|00003ae0|Trace|MSEndPoint.cpp(819): 0:EndPoint [email protected]:5482 destroyed statistics:
 
We are experiencing the very same thing. We have never had the audio (both ways) stop working before. It all started a few weeks ago when it updated to 15.5 Update 6. Since then the audio (both ways) stops working weekly (but not predictably). To fix it, I restart all 3CX services from the web interface and audio immediately beings working again. This definitely seems like a bug with Update 6, as nothing about our environment has changed except for the update. It happens about once per week, and it's very frustrating to our users.
 
I am also having the same issue since installing 15.5 Update 6. Our SIP provider is actually dropped into our network behind the firewall, therefore it is local lan to local lan connection. However, the calls still just stop providing audio. I have restarted the server completely. Seems to clear the problem for a while. But then it is reoccurring randomly several days later. I cannot determine how to replicate the issue and therefore cannot turn on logging to catch it.
 
Unfortunately, for me this has been going on for several months now. I installed the SBC because I had problems with STUN and it improved but did not go away. I am unsure of the resolution, but I certainly have a lot of angry end users.
 
Everyone here is aware 3CX changed the RTP port range right?
 
@Cody Fontaine

Don't think the RTP port range change necessarily affects you as it shoudln't change in the middle of a call. Your double NAT setup is bad so I'd pay the extra $20/month and get a static IP from Comcast as step one.
 
I am aware and have updated the range, however my problem occurred before the update to 6
 
Yep... was kinda of a general note to folks and I was in the middle of composing a more specific reply to you about that.
 
@
cobaltit said:
@Cody Fontaine

Don't think the RTP port range change necessarily affects you as it shoudln't change in the middle of a call. But your double NAT scenario is bad both because of double NAT and security so I'd pay the extra $20/month and get a static IP from Comcast as step one.
Click to expand...
What if I were to put the gateway into bridge mode and eliminate the double nat? Could that potentially provide a resolution to the issue? I would believe that the ports were getting messed up between the two.... Should have considered that myself.... Thanks for the suggestion though.
 
Well VoIP is 95% networking so if I were being paid to fix the problem the first thing I would do is eliminate the double NAT. I've also found that the Comast equipment in general gets pretty shaky under load and have solved a number of stability issues with them but simply using them as a modem and having the router behind do NAT. Honestly I don't like any of the equipment you have there :). The newer Cisco modems have always been problematic for me (except for one customer, not sure why) and I always had the guys pull a SMC out of the truck for me instead. Your RV220w is getting pretty old as well having been EOL'd almost 4 years to the day so that could be causing problems as well.
 
cobaltit said:
Well VoIP is 95% networking so if I were being paid to fix the problem the first thing I would do is eliminate the double NAT. I've also found that the Comast equipment in general gets pretty shaky under load and have solved a number of stability issues with them but simply using them as a modem and having the router behind do NAT. Honestly I don't like any of the equipment you have there :). The newer Cisco modems have always been problematic for me (except for one customer, not sure why) and I always had the guys pull a SMC out of the truck for me instead. Your RV220w is getting pretty old as well having been EOL'd almost 4 years to the day so that could be causing problems as well.
Click to expand...

Comcast took my smc I had and replaced it with this Cisco gateway and said that they were no longer providing the smc modems for the faster speeds. I should have thought of the double Nat myself...rookie mistake. The Cisco rv220w is in the process of being replaced. I recent took over the department, my predecessor never wanted to change anything, therefore the higher ups don't really understand why IT "suddenly" needs to such an investment... Haha.
 
@Cody Fontaine

Hah.. sounds like fun. Yeah forgot about needing something newer than the SMC for higher speeds. Guess you are stuck with the Cisco.
 
Other errors I am recieving:
10/09/2018 1:39:19 PM - [CM505001]: Endpoint Extn:321: Device info: Device Not Identified: User Agent not matched; Capabilities:[reinvite, replaces, able-no-sdp, recvonly] UserAgent: [3CXPhone for Android 15.5.344.492] PBX contact: [sip:[email protected]:5060]
10/09/2018 1:39:19 PM - [CM505001]: Endpoint Extn:319: Device info: Device Not Identified: User Agent not matched; Capabilities:[reinvite, replaces, able-no-sdp, recvonly] UserAgent: [3CX WebRTC proxy] PBX contact: [sip:[email protected]:5060]
10/09/2018 1:39:19 PM - [CM505003]: Provider:[1voip1] Device info: Device Not Identified: User Agent not matched; Capabilities:[reinvite, replaces, able-no-sdp, recvonly] UserAgent: [VYLmedia-SBCDAL] PBX contact: [sip:[email protected]:5060]
 
I removed the Double NAT last night. I will monitor and see if that helps external calls. However, I am still struggling with the internal calls losing audio also. My understanding is that the SBC should be handling these calls without reaching out to the PBX?
 
On a proper setup (SBC, current firmware, stock templates) then use the audio should route directly between the extensions unless you've turned on PBX delivers audio on the extensions. Have you tried running the echo test for a period of time or maybe just creating a dummy queue with some music? Then you can just test one extension into either of those and see if you still get audio issues.
 
Cody Fontaine said:
My understanding is that the SBC should be handling these calls without reaching out to the PBX?
Click to expand...
If 2 extensions are behind the same SBC then once the call is established the audio will stay between the extensions if you do not have the option PBX delivers audio enabled or you do not record calls. In those scenarios the audio will route through the PBX.
 
Status
Not open for further replies.

Latest Posts

Forum statistics

Threads
112,095
Messages
590,724
Members
165,071
Latest member
RobertDung
3CX Logo Footer Copyright © Sitemap Privacy
Top Bottom