NEWS FLASH Issabel to discuss multitenant

[krzykat]

I'm going to try to attend this, although I may need my wife around to help my questionable spanish translations.

Reservar Ahora
WEBINAR: Descubre el potencial de Issabel Multitenant
Jueves, Marzo 15, 2018
@ 11:00 AM (Hora Estándar del Centro)
Presentación de las funcionalidades de Issabel Multitenant y el modelo comercial destacando sus beneficios.

Issabel integra una solución de PBX Multitenant para proveedores de servicios, además de la gestión simple de PBX virtuales para un número ilimitado de compañías o empresas.



Al asistir a este webinar sin costo obtendrás:

• Respuesta a todas tus dudas e inquietudes sobre Issabel Multitenant.
• Descubrirás las principales funcionalidades de Issabel Multitenant, así como los beneficios que te ofrece.
• Conocerás la Arquitectura y el Modelo Comercial de Issabel Multitenant.

Issabel Multitenant en un sólo Nodo Virtual contiene todos los componentes requeridos.

Recomendado hasta 1000 extensiones y puede convertirse en una arquitectura multinodo en cualquier momento. La arquitectura multi-nodo usa varias Máquinas Virtuales para ampliar su capacidad.

Este tipo de despliegue permite escalar horizontalmente habilitando más nodos de PBX y mantener un sólo punto de administración.



Reserva tu lugar ahora, y sé parte de la evolución.

¿Quién debería asistir?

• Comunidad Asterisk
• Comunidad Issabel
• Integradores y proveedores de tecnología Open Source
• Usuarios, desarrolladores y otras personas apasionadas por la tecnología Open Source.

No tiene costo. Ingresa al botón en la parte superior o inferior y no te quedes fuera

 

[wardmundy]

Google Translation:

WEBINAR: Discover the potential of Issabel Multitenant
Thursday, March 15, 2018
@ 11:00 AM (Standard Time of the Center)
Presentation of the features of Issabel Multitenant and the commercial model highlighting its benefits.

Issabel Integrates to PBX Multitenant solution for service providers, in addition to the simple management of virtual PBXs for an unlimited number of companies or companies.



By attending this free webinar you will get:

• Answer to all your doubts and concerns about Issabel Multitenant.
• You will discover the main features of Issabel Multitenant, as well as the benefits it offers.
• You will know the Architecture and the Commercial Model of Issabel Multitenant.

Issabel Multitenant in a single Virtual Node contains all the required components.

Recommended up to 1000 extensions and can be converted into a multi-node architecture at any time. The multi-node architecture uses several Virtual Machines to expand its capacity.

This type of deployment allows horizontal scaling enabling more PBX nodes and maintain a single point of administration.



Reserve your place now, and be part of the evolution.

Who should attend?

• Asterisk community
• Issabel Community
• Integrators and suppliers of Open Source technology
• Users, developers and other people passionate about Open Source technology.

It has not cost. Enter the button at the top or bottom and do not stay out

 

[krzykat]

Ward, these guys seem pretty good, but think they could really benefit from some "English" exposure. I'm so busy now, but would like to help them with that. I'm planning to listen to the meeting with my wife around to see what they've got planned.

 

[wardmundy]

Both Elastix and Wazo seem to suffer from the language barrier unfortunately.

 

[billsimon]

Funny, according to 3CX, "multi-tenant is dead": https://www.3cx.com/blog/news/cloud-ready-pbx-multi-tenant/

Reading through the BS, I wonder why 3CX couldn't make it work, nor could Elastix MT.

 

[wardmundy]

@billsimon: Not sure I would classify many of the points in Nick Galea's article as B.S. Folks can read it for themselves and make up their own minds. My reading suggests it had less to do with "making it work" technically and more to do with providing a more secure platform for each organization. When you can set up robust VoIP platforms for as little as $1 a month, why would you want to put all your eggs in one basket??

Looking at the major security breaches over the last decade, the 3CX track record is pretty solid compared to some others, especially Elastix and you know who. Most of the compromises documented below would have brought down not only your organization's PBX but also those of all the other companies sharing the multi-tenant platform. So... one organization needs a component added and that component exposes a vulnerability that ends up bringing down the whole house of cards. That's a pretty dangerous design in my book, but to each his own.

"The total amount of detected vulnerabilities is very high. Luckily, the majority of the detected vulnerabilities are inside the administration control panel, such that attackers either need to steal a valid account or they have to trick an administrator into visiting a malicious website that triggers one of the critical vulnerabilities. For example, a remote command execution vulnerability could be triggered by a less critical cross-site scripting vulnerability. By chaining both vulnerabilities, the severity is increased drastically and can lead to full server compromise."

See: https://blog.ripstech.com/2016/freepbx-from-cross-site-scripting-to-remote-command-execution/

 

[billsimon]

The "BS" call comes because of 3CX's obvious spin on the current state of multi-tenant in order to sell their single-tenant product. Your arguments are right, but there's a certain threshold where scale changes the way things work. It makes sense to me to run 10 DO or OVH VMs for 10 tenants. But what about 50 or 100? Or 500? I recognize that there are tools to push out configs, patches, and upgrades. It's still a lot more work than updating the OS or software on a multi-tenant platform, once and done. Or setting a default config, or a provisioning template, or...

Anyway, I see the benefits of both approaches, but 3CX might want to put a giant asterisk (no pun intended) or three next to their claim that multi-tenant is dead. * In small deployments where each tenant is fairly unique. * May not apply to large hosted PBX services. * Each server licensed separately.

A subnet full of individual vulnerable VMs seems like an equivalent surface area for attack as a multi-tenant platform. And when the attack comes and has gone, it's a huge mess multiplied by 10. (Ask me and one of my clean-up clients how I know about this.)

Now the reason I'm going on about this is that I recently revisited the state of multi-tenant to see what's out there. Since 3CX is closed I don't know about their core, but for the major two open-source PBX contenders, Asterisk and FreeSWITCH, only one of them is built with multi-tenant capabilities at the core (FreeSWITCH). FusionPBX and Kazoo (GUIs for FreeSWITCH) manage this beautifully. As for Asterisk, doing multi-tenant with dialplan and device configs is not possible other than in a very hacky, incomplete way. If Issabel wants to go at it, their best bet would be to develop all of the user/config/routing/features layer with a proxy system like Kamailio or OpenSIPS and use Asterisk only for media functions. It's a ground-up redesign.

 

[wardmundy]

Sorry, but your examples are a little far-fetched and well beyond the capabilities and requirements of most of the Asterisk and 3CX platforms and organizations covered here. Certainly, there may be use cases for multi-tenant on some platforms (other than Asterisk and 3CX) under unique and limited circumstances, e.g. a single multi-tenant system for the U.S. State Department situated in hundreds of countries. I'm having a hard time coming up with many other examples. My point is those clearly weren't the target audience of the article you referenced. Nor would many agree that organizations with several hundred or more phones are small deployments.

 

[krzykat]

I actually do see the potential need for multi-tenant depending on what you are doing. Let us say you have 50 clients so you have 50 PBX's on VM's. It's nice because no single point of failure. But let's say you want to change your default call routes across the board. Would be nice to make one change instead of 50. But depending on the multi-tenant application, you still might need to. Likewise, let's say Ward comes up with a fancy new way to have Lenny 2.0 - would be nice to install it once instead of 50 times. There would definitely be pro's and con's to each approach, but I also agree that probably 98% of people on this forum - multi-tenant wouldn't even be a consideration.

 

[billsimon]

It's a good debate. I have worked in different environments and all of them make me mad at some point so I am not here to proclaim anything as truly wonderful. My opinion about Issabel is that they should not try it unless they can make it scale.

For as long as I have kept my eyes on the various VoIP PBX projects we talk about around here all the time, I have seen some interest in multi-tenant and lots of "creative" ways of squeezing three companies into a single PBX (custom contexts and such). There's demand, though it might not be well thought out.

 

[jerrm]

I read the article as mostly marketing BS. That doesn't mean some of the points aren't valid. At the end of the day, 3CX is discontinuing their multi-tenant product and trying to put a positive spin on the way forward. Nothing more, nothing less.

As for it's usefulness, there are advantages both ways, I can think of a couple of prior lives in M&A heavy environments where it would have been useful, but multi-VMs would still be light years better than the solutions at the time.