Asterisk and Linux to Build Secure VoIP Connection

Beave writes "Using Linux and the Asterisk PBX, it is possible to build a secure, cost effective VoIP (and traditional PSTN) PBX solutions. This article shows you how to take advantage of various hardware, software and tricks to accomplish this goal within a limited budget."

Comments

Shows you how?

[Dancin_Santa]

More like tells you in the most general of terms what they implemented.

Obviously what is going to be the real killer app is VoIP in a wireless setup. Instead of having a wall jack for your desk phone, it just hooks into the wireless mesh seamlessly.

I'm sure this has already been done. I'd love to see an article about it.

A view from the industry

[jaymzter]

From an enterprise viewpoint, that is a very large service base, asterisk is dead in the water until it can match the simplicity of the interfaces found on proprietary systems. This isn't a knock on asterisk as a technology solution, but the telcom admin of a large corporation isn't going to want to look at a text file to figure out his dialplan or use some arcane interface when on a more mature system he can use a simple command like 'display dialplan'.
I don't doubt many people have used asterisk as a voice solution for some companies, but not for any major companies and certainly not for any huge call centers. RTFA, a CIO would sh*t if you showed him snippets from some text file. Not to mention the questionable logic of running your voice system on a white box computer. It may be fine and dandy when e-mail is down for an hour, but five minutes without phones is a lifetime for any serious company. 5 9's is not a joke in the voice world and actually a rational expectation.
In other words, I support asterisk simply because I love open source, but don't kid yourself, right now it's just a hobby app (as seen from the enterprise)

Re:A view from the industry

[Stinking Pig]

Actually, it's fairly uncommon at large companies for the CIO to give a damn about how the thing is configured... that's your problem, Slashbot. Similarly, they wil indeed sh*t if they find it's running on a white box computer... but the choice of platform is your problem, not the software's.

I'm not disagreeing with you about Asterix's readiness or lack thereof, I don't really have an opinion. But I do have a lot of experience with CIOs going through the buying process, and I can tell you that they think in terms of services rendered, costs incurred, and risks undertaken. If you can provide the service required at a reasonable cost with minimal risk, they really don't care how you're going to do it so long as your solution passes the sniff test. White box hardware in a mission critical solution stinks to high heaven, but an open source app on good gear with a clear set of contingency and support plans has a fighting chance these days.

Re:What about Speex

[Anonymous Coward]

Yes - Yes you can.

limited budget indeed

[SuperBanana]

I purchased three Intel white-box computers for $800 each containing 2.6Ghz processors 512MB ram and 40 GB hard drives

Anyone who recommends greybox PCs with non-raid storage for a financial institution...even a small one with only three branches...is not thinking very clearly. If it's for a business-critical application like the phone system, they're categorically insane.

Folks- there's a reason those telco boxes cost lots of dough. They Just Work if they're left alone (in 7-8 years of working with telco equipment, 99% of the problems have been telco line provider problems; hardware failures are extremely rare). There are books upon books written with guidelines for what is considered telco grade, but the common theme is "keeps going, and if it breaks, it does so gracefully".

$2500 can, even for a small bank, be PENNIES ON THE DOLLAR when the system goes down for even a few hours. If you've got a Lucent phone system and a support contract, they find stuff before you do, and no matter what time of day- there's a tech on your doorstep in an hour if they can't remote in via the system's POTS admin modem.

You want a cheap phone system, you get what you pay for. It's remarkably irresponsible for the authors of that article to advocate Asterisk without mentioning that reliability and support pale in comparison to 'real' telco equipment.

Re:limited budget indeed

[Anonymous Coward]

While the authors were remiss in suggesting the systems they did there is no reason why an Asterisk based system can not be built with the redundancy and scaleability of a "real telco solution". Telcos and large suppliers have a vested interest in making telephony seem more difficult then it is, and frankly the cat is out of the bag. It is very possible to build a system with a very high level of reliability for a fraction of the cost of a commercial system. I've had a lucent phone system and a support contract and it isn't all that its cracked up to be.

The key to building these solutions is alot of redundancy both in the hardware and with clustering. Voip + Clustering gets you very close to telco grade. Support is really the biggest issue, but there are a growing number of companies providing it, and if its set up well you won't be needing much of it. I take issue with the parent poster disparaging these open source systems as somehow inferior to products from Avaya, Nortel and others. Its just not true anymore. The only area where these vendors have any competetive advantage is in their interface designs, and as convoluted as Asterisk is I'll take it anyday over Avayas windows management software.

Re:limited budget indeed

[isometrick]

There's a company that banks their proven reliability on the redundancy of their software. Not hardware.

I think they are doing pretty good.

Trans-Atlantic VoIP

[Evanrude]

A friend of mine, who works for a UK telecom and my company, Axigent, have setup a connection between our two asterisk systems that has proven fairly reliable and "secure". I would say with everything we've gone through to make the connection functional, the author of this column left out a lot of the details as far as full implementation of an Asterisk PBX. A helpful site, or at least one of the more helpful sites I've come across is the wiki at www.voip-info.org, which the author neglected to reference in his article. Knowing someone that works at a Telecom is a plus, I think the cost from both ends as far as equipment has been fairly minimal and the return on the time invested as far as learning what VoIP is capable of has been huge. All of the calls that are made back and forth have been clear. It's pretty impressive to call overseas at no charge.