Slashdot Mirror
Pro-Active VoIP Management Solutions?
Adeptus_Luminati asks: "I've been running a 1000 user Mitel VoIP phone (to the desk) network which encompasses 20 buildings glued together by our Telco's _private_ fibre backbone (no Internet involved here). Once in a while we have voice quality degradation issues caused by excess latency, jitter, bandwidth saturation, QoS mis-configurations, and so forth. I've been using Ixia Chariot software to simulate VoIP calls over the WAN between our various offices and collect data of the problems, but this is only useful AFTER the problem is reported by our users, and after I am lucky enough to be around and catch the problem happening in real time; otherwise, I have no way of proving to our Telco that there IS a problem. What solutions have other network admins come up with to pro-actively manage similar private VoIP networks?"
"I am looking for some sort of solution to allow me to pro-actively monitor or simulate 24/7 VoIP calls between offices and then report back to me immediately when certain thresholds of voice quality degradation have been exceeded and accumulate significant info that I can forward my Telco and get them to deal with the problem, right away. FYI, bandwidth is free on my office WAN links, we're mostly 100Mbit fibre, and we have QoS from end to end (except small parts of the telco backbone)."
Computer #1 in one building, #2 in another.
Cron job:
Computer #1 voice-calls computer #2 and plays a complex and long sound.
Computer #2 records the sound it received.
Computer #2 compares the sound it received with the original file.
Log errors; if error-rate > x, page you, sleep short time, repeat cron job.
Simple, ain't it?
It's better to be the foot on the boot than the face on the pavement. ~~ tkx Kadin2048
Pro-Active VoIP Management Solutions?
You're going to hell.
there is also the option of turning down the audio quality between buildings. (ie, 128Kb stream inside the building, 64kb stream between them.) While slightly more noisy, it still works, and uses less bandwith. I know with our old Cisco VOIP at my old job, department to department calls were low bandwith, and customer calls were setup for highest bandwith. (clearest)
What are we going to do tonight Brain?
The Voice Option is a value-added package that integrates with InfiniStream Network Management to provide additional insights into voice- and video-over-IP converged traffic. Voice-over-IP (VoIP) Experts automatically detect and help resolve key problems seen on VoIP networks--jitter, packet loss, packet-sequencing errors, and latency. These VoIP Experts and call-tracking capabilities, along with the traditional Expert system, help ensure successful VoIP network rollouts while maintaining "toll-quality" voice and high-quality data for all users.
The product URL is here
They make a couple of versions. The last time I looked, the 1 TB version was around 25K and the 4 TB version was around 95K. I didn't buy one, but it was a fun toy to play with.
Still, with a plan, you only get the best you can imagine. I'd always hoped for something better than that. -CP
to reply to my post, This looks interesting NQMS
What are we going to do tonight Brain?
It sounds like you're talking about monitoring your network for application performance and watching for telltales that precede degredation. You might check out a product from this company:
Because that's exactly what they do :)
Unlike a general packet sniffer or network monitor, they aim exactly at your kind of problem.
Disclaimer: until I entered the glorious realm of academic programming, I was employed by NEXVU, and I still have stock and stock options. Even though they no longer benefit from my obvious genius, I think they have a great product. If enough VoIP administrators agree, then those options may be worth something in the future!
Check here.
You shoulda stayed with the olde way of doing things. This new fangled VOIP cr*p is nothing but a load of POS foisted upon yee by sales and marketing droids to give them a reason to exist. And sell yee more cr*p.
If you have to prove to them that their is a problem, they aren't likely to do anything about it when you do come up with evidence.
Sorry to burst your bubble, but unless you call your sales rep and threaten to leave, your not going to get anywhere.
symetrix. We are building a religion, a limited edition.
His solution? He got his board of directors to approve the purchase of some wifi radio equipment, which they mounted on nearby towers. I am not a hardware or radio guy, but this was not Linksys crap that I run in my home. He got some professional stuff. Each office had LOS to a local tower, and the towers to each other. Last I heard, they are running all of their voice and all of their data over their new links. Routers at both ends are configured for QoS, and thing are running very well. The cost of the equipment has already been paid for with the savings since what they pay for the towers is a fraction of the cost of the circuits they were running between offices. They maintain a few landlines that the phone systems on each end can use in the case of emergency to route voice traffic, and I believe he also has a couple of redundant DSL lines for data.
Great ideas often receive violent opposition from mediocre minds. - Albert Einstein
I don't know of an existing tool for this, but you want to measure "known traffic" across the route and report when it degrades.
Setup an "application ping server" on the far end. This will be a C program that posts a UDP listen on a port in the RTP range. When it gets an inbound packet, it returns exactly that same packet to whom sent it to them. This needs to be in C (or similar) because the latency needs to be very low. It should also run on a very low utilization server.
On the near end, write a similar program that sends a UDP packet to the far end and measures round-trip delay. If you want to really act like a VOIP call, send 200 byte packets at 50/sec (or whatever the SIP/RTP payload is for your particular codec).
The question is then how to analyze the data. I would probably have the program spit out peak/avg "numbers" for latency and packet loss into a text file, perhaps at 1/minute. You could then have MRTG (or your favorite grapher) read those and draw you pretty stuff. Once you see the types of degredation that actually matters, you can program in your thresholds to page/email/whatever you.
Things to remember when setting this up. Make sure the traffic is handled by the routers exactly like your VOIP packets. Run on the same UDP ports with the same QOS tags. In other terms, make your test traffic indistinguishable from the actual voice traffic (at least from a router's/switch's voicepoint).
Of course, someone probably has a package that does this. If not, build it and open a sourceforge project.
Expensive option:Empirx Hammer XMS. It does all of the above with a nice web interface plus it gives you RTP quality metrics like r-factor and MOS. It's not cheap, but I've used and it does a good job (it is basically a SuSE Linux box with some networking gear running their network monitoring software).
All of the above I have tested only with SIP/RTP traffic. If you youse MGCP or H.323, I can't personally vouch for either of the above solutions, though both support them.
There's no place I can be, since I found Serenity.
http://www.prognosis.com/ Also, consider IP SLA (Cisco) GrpA
Enjoy science fiction? "Turing Evolved" - AI, Mecha, Androids and rail-gun battles. What more could you want?
Fibre = Fibre Channel protocol as used in a SAN
Fiber = Fiber optic (as in the physical cabling)
Sorry. It's an annoying habit of mine.
Give Smokeping a try. Smokeping:ping::MRTG:bandwidth.
i ng/
http://people.ee.ethz.ch/~oetiker/webtools/smokep
It shows latency, loss, and jitter in a combined easy to read graph. After using it for a while, you can spot many normally invisible network anomolies on these graphs long before they become visible to users. They're also great for post mortem analysis.
They don't have anything to do specifically with VoIP, but I think they're invaluable tools for any network admin.
http://edgewaternetworks.com/
It's specifically designed for VOIP quality monitoring.
And as a disclaimer, I do some work for the company.
100mbit backbone are you joking?
QoS partially implemented, are you joking?
Think about it, 100mbit is just about adequate for the Store and Forward world of data packeting and is hardly adequate for the real time packet delivery world of Voice. Unless its just you and that hot admin assistant exchanging essential personal data!
Chariot does both monitoring and call load simulation... exactly what you want.
I'm in charge of network assessments for a very large voip hardware manufacturer... we've used this tool to do what you're describing.
Give tech support a call and get them to walk you through it. It's a great tool.
Make sure you've got QoS properly set up on all your devices too, regardless if it's across the internet or not. You still need QoS!