This has evolved over time as unified communications has evolved. To really be accurate it should be done with a solution that uses packets as its data-source. This is really a not a good play for NetFlow or SNMP polling and traps. Some log companies will also throw their hat in this ring by collecting CDR {Call Data Records} from call manager servers on their opinion of the quality of the call by something called MOS {Mean Opinion Score}.
Each manufacturer/vendor will have some kind of tool to investigate at least the signalling {SIP, Skinny, etc} and maybe the signalling {RTP, etc} but that is also the constraint, many don't support other vendors. Most vendor-agnostic tools from NPM {Network Performance Managers} products can at least do Jitter and Out of Sequence and Packet Loss on the media {RTP} streams of a call, which network teams need to prove their innocence. Example vendors would be Riverbed, Extrahop, Viavi, NetScout. But these same vendors lack the Signalling correlation to the call. Then, today there is the SaaS aspect too. Like WEBEX, Zoom, 8x8, Teams, etc. and this traffic the media {RTP} is usually encrypted and often the signalling {SIP} is encrypted as well.
The only vendor I have tested and monitor as of today, right now, is NetScout. Their product is vendor agnostic, supports all versions of signalling and media, including video. Also, they have the ability to decrypt the HTTPS traffic going out to any SaaS vendor, even TLS 1.3. It even has the ability to "listen" to the calls and flag things like echo, tinny like sounds, soft voice, loud voice, background noise, and correlate to factors like QoS mapping, Jitter, Loss, OOS, etc. It also monitors and alerts on One Way Calls, DTMF mistakes, Voice Mail, Conference Call stitching, Call forwarding following, path latency and network error hop by hop. A very long list of other features. They even know what each CODEC can tolerate in the aspect of network errors and call quality as far as sound goes. This means if the CODEC can absorb the issues it will log and baseline the behavior but not Alarm on it unless you want it too.
The last point I will make is really about privacy laws worldwide. Replayability in any investigation only or investigation & monitoring solution has to be tightly controlled and logged and if that is not possible then do not save the packet payload on the media.
Search for a product comparison in Network Monitoring Software
NetVoice from NIKSUN is the best. As NIKSUN has Full Packet Capture technology coupled with Google, like a search engine with user-friendly GUI and reporting abilities.
Network Engineer at a government with 1,001-5,000 employees
Real User
2020-06-25T19:56:51Z
Jun 25, 2020
You can use the IP SLA feature running on Cisco routers. As so, you can define many protocols to monitoring. You have to confogure a router as IP SLA sender and many routers or devices as IP SLA responders. Depending of the feautures of the responders, you can see ICMP, UDP, TCP responders. If you have VoIP responders, you can monitor also the jitters and the MOS values.
All of this results are loaded in the IP SLA sender router. The you can see theses statistics (history, graphs) with monitoring tools, as PRTG or WUG. Are you working with Cisco devices? If yes, this is a great and powerful way.
Works at a comms service provider with 11-50 employees
User
2020-06-24T21:47:20Z
Jun 24, 2020
Nagios XI is the commercial, easier-to-use version of the Nagios core open source version, but still has the extensibility of core. Tends to be very price competitive too.
PRTG is the best that I have used ever, Check its sensors and If you didn't find the sensor that you are looking for you can create your own one meanwhile you can request it from the development team.
My recommendation using Nagios XI. they easy to implement and easy to administrating.
Nagios XI also have the Plugin for VoIP
Nagios Plugin to check Call Quality in SIP VoIP (compatible with checkmk, etc) sipnagios implements the Nagios plugin API for monitoring and performance data. sipnagios.c is a modification of the original siprtp.c sample in pjproject distribution
This has evolved over time as unified communications has evolved. To really be accurate it should be done with a solution that uses packets as its data-source. This is really a not a good play for NetFlow or SNMP polling and traps. Some log companies will also throw their hat in this ring by collecting CDR {Call Data Records} from call manager servers on their opinion of the quality of the call by something called MOS {Mean Opinion Score}.
Each manufacturer/vendor will have some kind of tool to investigate at least the signalling {SIP, Skinny, etc} and maybe the signalling {RTP, etc} but that is also the constraint, many don't support other vendors. Most vendor-agnostic tools from NPM {Network Performance Managers} products can at least do Jitter and Out of Sequence and Packet Loss on the media {RTP} streams of a call, which network teams need to prove their innocence. Example vendors would be Riverbed, Extrahop, Viavi, NetScout. But these same vendors lack the Signalling correlation to the call. Then, today there is the SaaS aspect too. Like WEBEX, Zoom, 8x8, Teams, etc. and this traffic the media {RTP} is usually encrypted and often the signalling {SIP} is encrypted as well.
The only vendor I have tested and monitor as of today, right now, is NetScout. Their product is vendor agnostic, supports all versions of signalling and media, including video. Also, they have the ability to decrypt the HTTPS traffic going out to any SaaS vendor, even TLS 1.3. It even has the ability to "listen" to the calls and flag things like echo, tinny like sounds, soft voice, loud voice, background noise, and correlate to factors like QoS mapping, Jitter, Loss, OOS, etc. It also monitors and alerts on One Way Calls, DTMF mistakes, Voice Mail, Conference Call stitching, Call forwarding following, path latency and network error hop by hop. A very long list of other features. They even know what each CODEC can tolerate in the aspect of network errors and call quality as far as sound goes. This means if the CODEC can absorb the issues it will log and baseline the behavior but not Alarm on it unless you want it too.
The last point I will make is really about privacy laws worldwide. Replayability in any investigation only or investigation & monitoring solution has to be tightly controlled and logged and if that is not possible then do not save the packet payload on the media.
NetVoice from NIKSUN is the best. As NIKSUN has Full Packet Capture technology coupled with Google, like a search engine with user-friendly GUI and reporting abilities.
You can use the IP SLA feature running on Cisco routers. As so, you can define many protocols to monitoring. You have to confogure a router as IP SLA sender and many routers or devices as IP SLA responders. Depending of the feautures of the responders, you can see ICMP, UDP, TCP responders. If you have VoIP responders, you can monitor also the jitters and the MOS values.
All of this results are loaded in the IP SLA sender router. The you can see theses statistics (history, graphs) with monitoring tools, as PRTG or WUG. Are you working with Cisco devices? If yes, this is a great and powerful way.
I have seen Nagios, PRTG and ManageEngine
These are all appropriate tools. Just adding my own experience as having the opportunity to have worked with all.
Nagios , easy to implement, agents easy to install, interface management for alerts: good
PRTG , easy to implement, agents very easy to install, interface management for alerts: good
ManageEngine is more modular to implement, agents easy to install, interface management for alerts: very good
Using LogicMonitor, you can monitor network equipment and servers, I think this depends on your equipment,
Nagios XI is the commercial, easier-to-use version of the Nagios core open source version, but still has the extensibility of core. Tends to be very price competitive too.
PRTG is the best that I have used ever, Check its sensors and If you didn't find the sensor that you are looking for you can create your own one meanwhile you can request it from the development team.
https://www.paessler.com/prtg
I kindly suggest Manage Engine Opsmanager.
You can use PRTG Network Monitor. It has the built-in ability to monitor VoIP between end points.
Feel free to contact me in case you want to discuss more on that.
My recommendation using Nagios XI. they easy to implement and easy to administrating.
Nagios XI also have the Plugin for VoIP
Nagios Plugin to check Call Quality in SIP VoIP (compatible with checkmk, etc) sipnagios implements the Nagios plugin API for monitoring and performance data. sipnagios.c is a modification of the original siprtp.c sample in pjproject distribution
You can have it at github.com/gmaruzz/sipnagios