Difference between adtran and polycom phones

[natem]

  There seems to be a difference in the configuration or call handing with our polycom phones and adtran phones. Our adtran phones are able to call extensions on our other adtran system but our polycom phones on one side are not able to hear the audio when calling the other system. I've tested both types of phones on the same port and had no issues with the adtran phones.

  Has anyone had this issue before or could someone recommend the settings/ftp files that need to be compared?

  I've spent a lot of time on this issue and have compared polycom files with between our other 7100. I've matched them up and they seem to be good.

All polycom phones are IP332 and adtran are 712.

[natem]

I found that enabling the polycom phones as simple remote phones fixed the one way audio to the other nv7100 on the LAN. The polycom phones on the other nv7100 are not enabled but still work.

I don't have a the firewall enabled on either NV7100 that are connected via LAN so I'm wondering why this fixed the audio problem. Anyone know what the remote phone setting changes in the header or why this works?

The adtran phones don't require remote-phone. Does anyone know why a polycom phone requires this setting but not an adtran phone?

[Anonymous]

natem,

You should not have to set the Polycom phones up as simple remote phones to make this work.  My first suggestion is to check and make sure that media anchoring is off globally.  It should be set to no ip rtp media-anchoring in the configuration.  If you still have problems after that it would be helpful to know the topology of how these units are connected.  It could be something related to a firewall or NAT device causing the problem, but I would expect the ADTRAN phones to operate the same way.

Can you recreate the issue and get a debug voice verbose, debug sip stack messages, and a debug isdn l2-formatted from the site with the problem?  You can omit the isdn debug if you are not using a PRI.  Once you have that debug please make a .zip file with it and the current configuration of both units and upload it to our FTP server with the instructions below:

Open Internet Explorer web browser
Type the following URL:  ftp://ftp.adtran.com

Press the Alt key, click View, and then click Open FTP Site in Windows Explorer

Double-click the "Incoming" folder
Drag and drop files from PC into the Internet Explorer window

Reply to this post with the file name so we can retrieve it

Thanks,

Matt

[natem]

GreekPeak.zip

Thanks a lot Matt.

The firewalls are turned off on the two 7100s as they are both behind a 908e on the same LAN. I ran the no ip rtp media-anchoring on the two 7100s. The RTP media-anchoring wasn't enabled prior either.

[Anonymous]

natem,

Thank you for the information.  Can you supply another debug with debug voice verbose and debug sip stack messages both enabled at the same time for each unit?  Ideally I would like to have a debug file from each unit for the same failed call containing the output from both of those debugs mixed together.  Sorry, I did not clarify that earlier.  Also, which user experiences the one way audio (6119 or 6400)?

Thanks,

Matt

[Anonymous]

natem,

Thanks for supplying the new debugs. 

The first thing I would recommend is upgrading the Polycom 331 firmware.  It is currently running 3.1.3 and should be on 3.2.7.  The phone firmware requirements are outlined in the release notes for each version.  You are running AOS R10.6, so this can be found on page 6 of the .  The Polycom firmware can be downloaded from this page on our main website:

http://www.adtran.com/web/appmanager/portal/Adtran?_nfpb=true&_pageLabel=waverunner_page_support_sof...

This may very well fix the issue, so please try this first.  If you continue to have problems afterwards I would need the questions below answered:

You mentioned the problem was calling from a Polycom phone to another phone on the remote 7100, and that this works if the same call flow is recreated with an ADTRAN phone.  The debug indicates extension 6119 is calling 6400, which is the voicemail extension number on the remote 7100 instead of another phone.  Can you confirm if 6119 can call an extension that goes to another IP phone off the remote 7100? 

Both units are on the same LAN but they have different default routes.  One is to 192.168.0.23 and the other is to 192.168.0.15.  I assume that the Polycom and ADTRAN phones off of each 7100 are using DHCP to obtain their default gateway so they should be the same (the voice vlan IP of each local 7100), but wanted to make sure these phones are not manually programmed.  To confirm there is not a routing issue can you plug in a PC off of each 7100, set it to be in the voice vlan to mimic a phone, and then ping  the a phone in the remote voice vlan as well as the 7100 sip trunk address?

For comparison purposes it would also help to see a debug of this working with the ADTRAN phones and if possible a packet capture when the issue is recreated again from the port the phone is connected to with matching debugs on each unit again.

Thanks,
Matt

[natem]

It appears that the updated firmware fixed the one way audio issue that was happening with the polycom phones. I thought that the phones were running this version since I had put the files on the cflash but I didn't do this correctly.

I have noticed that there is an issue with system audio not playing or being heard when calling from a polycom to the other 7100. By system audio I mean that there is no audio with the following, ring tone, voicemail greetings, auto attendant, call queue, and hold music. I removed the simple remote phone option. I was using the voicemail extension as a test and this caused confusion since the other audio appears to be fixed by the phone firmware update. If I set the simple remote phone option it will allow the polycom phone to hear the system audio.

  I did a quick comparison between a new adtran phone call debug and a polycom phone on the far end 7100. I noticed that a=silenceSupp:off - - - - is present along with a=fmtp:18 annexb=no for the adtran phone. I don't know if this was a configurable option somewhere or if the polycom phone doesn't support this option. A guess is that RTP traffic isn't being sent. Is VAD configurable? The comparision is shown below:

far end debug with adtran phone:

19:31:52.907 SIP.STACK MSG         v=0

19:31:52.908 SIP.STACK MSG         o=MxSIP 0 1039271676 IN IP4 192.168.5.30

19:31:52.908 SIP.STACK MSG         s=SIP Call

19:31:52.908 SIP.STACK MSG         c=IN IP4 192.168.5.30

19:31:52.908 SIP.STACK MSG         t=0 0

19:31:52.908 SIP.STACK MSG         m=audio 3000 RTP/AVP 0 18 101

19:31:52.908 SIP.STACK MSG         a=ptime:20

19:31:52.909 SIP.STACK MSG         a=sendrecv

19:31:52.909 SIP.STACK MSG         a=silenceSupp:off - - - -

19:31:52.909 SIP.STACK MSG         a=rtpmap:0 PCMU/8000

19:31:52.909 SIP.STACK MSG         a=rtpmap:18 G729/8000

19:31:52.909 SIP.STACK MSG         a=fmtp:18 annexb=no

19:31:52.909 SIP.STACK MSG         a=rtpmap:101 telephone-event/8000

19:31:52.910 SIP.STACK MSG         a=fmtp:101 0-15

19:31:52.910 SIP.STACK MSG    

19:31:52.912 TM.T04 01 SipTM_Idle           rcvd SIP call-leg request: INVITE

19:31:52.913 TM.T04 01 SipTM_Idle           call-leg -> Offering

19:31:52.913 TM.T04 01 SipTM_Idle           State change      >> SipTM_Idle->SipTM_Trying

with polycom phone:

00:50:17.730 SIP.STACK MSG         v=0

00:50:17.730 SIP.STACK MSG         o=- 1359611419 1359611419 IN IP4 192.168.5.14

00:50:17.730 SIP.STACK MSG         s=Polycom IP Phone

00:50:17.730 SIP.STACK MSG         c=IN IP4 192.168.5.14

00:50:17.730 SIP.STACK MSG         t=0 0

00:50:17.731 SIP.STACK MSG         a=sendrecv

00:50:17.731 SIP.STACK MSG         m=audio 2236 RTP/AVP 0 18 127

00:50:17.731 SIP.STACK MSG         a=rtpmap:0 PCMU/8000

00:50:17.731 SIP.STACK MSG         a=rtpmap:18 G729/8000

00:50:17.731 SIP.STACK MSG         a=rtpmap:127 telephone-event/8000

00:50:17.732 SIP.STACK MSG         a=fmtp:127 0-15

Thank you for looking into the previous debugs and for your help. If I have to enable simple remote phones I think that this is an acceptable solution but was worried there was a bigger problem.

[Anonymous]

natem,

I'm glad to hear the new firmware fixed the issue.  For the system audio problem I would need to see the same debugs.  It would help to see them not working with a Polycom phone with an accompanying packet capture if possible, and then working with an ADTRAN phone for comparison purposes (no packet capture necessary for the working case).

Thanks,

Matt

[natem]

I've uploaded GPtoVoicemailDebug-PacketCapture.zip.

I noticed a change in the symptoms after I restarted both 7100s. The adtran phones are now behaving the same as the polycom phones. When calling from an adtran or polycom I don't hear the voicemail greeting, auto attendant, or call queue music. I upgraded the firmware on the adtran phones to match the supported firmware in the documentation. The upload contains the debugs from both units and a packet capture from the far end unit. I let the call in the uploaded capture go to voicemail. I don't know if the correct debug is being used to troubleshoot this issue. If the phone is answered audio works fine so I don't understand why other audio that is played from cflash doesnt come through properly. Frustrating problem.

Thank you for your work thus far Matt.

[natem]

Is it possible to get further help by using PSVs for this issue?

[Anonymous]

natem,

Sorry for the delay.  The debug appeared normal, and the root cause of the issue is the topology/routing setup.  You need to have a different voice VLAN for each 7100 that the attached phones and the local SIP trunk use.  Most importantly you should not have a direct connection to the remote voice subnet by configuring it as a layer 3 VLAN on the local unit.  Instead traffic to the remote voice subnet should be routed through the default route.  This is important as it dictates the address that is used to send and receive media (RTP). With that caveat in mind, I would suggest configuring each 7100 to have its own local voice subnet and make sure the remote voice subnet is not also configured on the local unit.  For example, you can have 192.168.5.x be on one 7100 and 192.168.6.x be on the other, but not both on the same 7100.  You will also need to reconfigure the SIP trunk addresses to match this new setup by changing the sip-server primary address to be the remote voice VLAN IP address.  A very stripped down sample configuration is shown below:

7100 A

--------

!

interface vlan 2

  ip address  192.168.5.1  255.255.255.0

  media-gateway ip primary

!

!

voice trunk T07 type sip

  sip-server primary 192.168.6.1

!

7100 B

--------

!

interface vlan 2

  ip address  192.168.6.1  255.255.255.0

  media-gateway ip primary

!

!

voice trunk T07 type sip

  sip-server primary 192.168.5.1

!

Thanks,

Matt

[natem]

I've tried changing to the configuration above and found that I get a busy signal. In the sip debug there is a sip 2.0/not found message. I think you have narrowed the issue down correctly with the topology/routing. I tried changing the topology by going directly to the other 1544 so that there is only 1 1544 between the 7100s. I also tried creating static routes but didn't have success. I've placed an order for some PSVs in hope that I can get my topology and routing corrected.

Thank you for your help in narrowing down the issue. I will post the final resolution.

[Anonymous]

natem,

I would like to see if I can help by looking at the config files for your two 7100s and your 1544. Can you put them in a zip file. Open windows explorer and go to ftp.adtran.com adn then open the incoming folder and put the zip file in there. Let me know what the file name is.

-Mark

[natem]

GPConfigs2.zip

[Anonymous]

Natem,

Where are the phones located at each location?

Hope Lake

Is it only on the 7100

Greek Peak

7100 only or also the GP1544P?

[natem]

Hope Lake

The 1544 has a 924 attached to it for some of our analog phones.

Greek Peak

The 1544P has phones going through it as well.

[Anonymous]

natem,

You said there was only one 1544 inbetween the 7100s but the drawing you sent has two 1544 between the units. Can you resend an updated network diagram with correct network addressing. I need an updated version with the updated address that you did per Matt's recommendation. This will help visualize your network.

Thanks,

Makr

[natem]

The diagram is updated with our current setup. I bypassed the GP1544 and connected the GP7100 directly to the HL1544 by using another fiber line as a test. This was out of desperation after trying the config as suggested by Matt. It was never suggested to try bypassing. I did this to see if it would help at all.

[Anonymous]

Natem,

Can you tell me which ports on each device are connected to each other?

[natem]

GPWC7100 uses port giga-eth 0/2 to connect to GP1544P

GP1544P uses port giga-swx 0/24 to connect to GPWC7100

      connects to HL1544F via giga-swx 0/28

HL1544F uses port giga-swx 0/28 to connect to GP1544P

       connects to HLCCNV7100 via giga-swx 0/23

HLCCNV7100 uses port giga-eth 0/2 to connect to HL1544F