We are still running Call Manager 4.1 and we currently experiencing static for Music On Hold. Every .wav file that we play has static on the line when people are put on hold. What would cause the static on MoH? Especially for all the .wav files. We didn't have problems in the past, so I know the .wav files are the correct format and codec used. Is it possible the system is overloaded with calls in the queue, or just the mere fact the system is overloaded?
Can you please check if the MoH issue is present for internal IP-IP calls as well?
If yes, then try restarting the IP Voice media streaming application service to begin with.
If not, then the gateway routing the calls to the pstn needs to be checked for any misconfiguration / qos / dsp etc related issues.
Yes, the DSP resources are configured from the installed PVDM's. To check if the issue is due to oversubscription you need to verify if the issue is reported only during peak hours. If it is being seen during non peak hours as well then it could be an ios or dsp related bug. A router reload or ios upgrade can be tried first in that scenario.
You can use the DSP calculator tool to check the number of dsp's needed for your setup.
We have 4 T1 PRI's where we can take a total of 92 simultaneous calls. This morning I tested when we had a total of 27 calls, and still heard static where at times it would hold longer than others. So I'm already hearing static and we haven't even hit deep into the day where usage will increase. Here's details about our router.
Cards installed: VWIC-2MFT-T1-D1 VWIC2-2MFT-T1/E1
4 PVDM2-64 moddules installed.
The VWIC cards were installed a couple years ago by another engineer, do you think those cards are going bad? Or is it wrong to use 2 kinds of VWICs series like that? It has gotten progressively worse in the past few months. I believe we are covered per the PVDMs installed on the router for oversubscription, would you agree?
I did indeed reload the router, with no change. I'll need to obtain a new IOS version, but what bothers me is the fact that it was working at one point. No changes were made, yet we are now facing this issue.
I'm thinking about upgrading the router, and getting a 3845 series and moving over the PVDM modules, and VWIC cards. This router would be running IOS 15.1 I would need to get a purchase agreement from upper management, but at the same time I'm wondering what if we get the same results with the new router? I know that's kind of a hard assessement, and we'll only know if we test, but I still can't help but wonder what caused this issue. It is running a pretty old IOS version, so that is something to consider.
What do you think mgogna, move forward in purchasing the new router?
I appreciate your time and effort in helping.
I would upgrade the IOS on existing gateway and test. Once the IOS is upgraded the issue may go away, else we can still take a capture and some debugs from the gateway to find what could be wrong. At least we would have ruled out IOS and DSP as a possible cause of issue before collecting debugs / captures.
I attempted to disable the MoH where it doesn't play music anymore, and simply beeps every 10 seconds. I did this by setting the Run Flag on both MoH servers to No, which did indeed disable the music on all phones, but not for the user calling into the call center. Doesn't the script reference the same MoH server?
*edit* After doing some research, looks like the MoH is simply referenceing a .wav file. I'm confused how this works for MoH. The script is simply referencing a .wav file so it appears that it's not even going through the Call Manager MoH server. When I copy and play that .wav file it plays just fine. Does this change the issue and path of troubleshooting?
I wanted to ask you mgogna, how would you go about doing the capture and debug for this issue? I'm curious what commands are needed to troubleshoot this voice issue.
In the past I've used the debug isdn q931 only.
You can collect the following:
debug voip rtp connection
show voice dsp
Also, check for any interface drops, controller errors etc.
I see input errors on two of the 4 serial interfaces, along with about 13 CRC on those two interfaces. The other two serial interfaces are clean.