USE NAGIOS XI OR OTHER TO TRACK DOWN VOIP ISSUE

[vcissgroup]

I just implemented NagiosXI, Logger, Fusion, and Network.

Networking from the perspective of new phone system and QoS – not my area of expertise. Yet, I have a customer that needs help and part of my job is to ensure that these changes do not impact production. Which – has already happened.

Prior to my involvement, a 3rd party vendor made a change that caused a Spanning Tree event. The vendor is Cisco certified. Familiar with the “phone system”.

The AVPN is a managed network, the phone system is “managed”, and the customer dropped dollars for the third-party to make the necessary changes to the LAN equipment as well. All Cisco. Mostly 3750’s.

The network prior was not managed, earlier generation MPLS. Upgraded to AVPN. Some circuits higher CIR rates (my recommendations).

The “old phone system” – working fine. New phone system…. Static. Calls dropped. Voice mail complaints. (short list).

There is more but I don’t want to reveal too much in the way of who or where. This is not the first outage caused. The managed circuit has FIVE 9’s and was down for almost 4 minutes last week – one outage. No ticket generated. No alert generated. Customer of third party had to report it. (Isn't 5 9's total of 5 minutes and 39 seconds for entire year)?

My customers are hearing a muffling sound like talking through a pillow. 50% of the time in some cases. It is claimed being a loss of 200K per day.

Between SNMP and Netflow and Logging would that identify the issue?

[tmcdonald]

I'm going to move this out of the Nagios Ideas forum since this isn't related to discussion of future features or anything like that. Since it is more of a network issue I am moving it to the Network Analyzer section.

In regards to the question:

Between SNMP and Netflow and Logging would that identify the issue?

Nagios itself does not identify the root causes of issues in most cases. The purpose is to alert you when things go wrong so they can be addressed sooner. For example, you can be alerted when CPU usage spikes, but Nagios can't know that the spike was due to a hung thread in a webserver process. So in your case, Network Analyzer would be good for showing where your traffic is going (source and destination), and you could use Log Server to collect all the logs from your devices, but I don't think anything will be able to tell you "The voice quality is declined because XYZ".

[eloyd]

No Nagios products, that I'm aware of, would really help you out here. Which is hard for me to say, because we make 100% of our money selling and supporting Nagios products as a third-party partner of Nagios Enterprises.

Having said that, it sounds like what you really need is a good SIP capable protocol analyzer. In a pinch, you can use Wireshark (actually, it's a great tool for that), and there is even some previous presentations on using Nagios with Wireshark, though not in a VoIP environment that I'm aware of. You might want to Google "Wireshark Nagios" and see what you get on the "videos" tab.

At the risk of going way off topic, VoIP voice quality is degraded by two things: jitter and packet loss. Those can be caused by a lot of things, but if your codecs are consistent and your network is properly tuned, then you really have nothing left to blame but equipment.

[dwhitfield]

At the risk of taking this back into XI-land, I didn't see any reference to "our" telephony plugins at https://exchange.nagios.org//directory/ ... /Telephony in https://support.nagios.com/forum/viewto ... 999#bottom . Although you still need to set up the checks to check the appropriate metrics, there might be something in there that can help diagnose your issue.

I see you have a 2pm CT Quickstart, so just let us know if you still have any questions after that.