Developers needed for an asterisk project

Hi

I’m not a developer. Our organization has conducted a thorough cost comparison for different hosted PBX providers. I recently came across Asterisk and after reading more I sense there is an approach which we ought to also explore in having a solution built and configured for us.

I would appreciate some guidance with the following questions:
1- Can anyone direct us to developers or teams which have experience in deploying a solution using Asterisk?
2- Kindly comment on what budget (deployment and maintenance) we should expect to have a solution done?
3- What common drawbacks exist when pursuing this route?

Again, help is sincerely appreciated and pardon the lack of developer background.

There would need to be several long discussions to develop something for your specific interest. Since you’re not a developer, you may want to look at something like FreePBX. It’s asterisk-based and comes with a user interface already built. I wouldn’t use it for a large-scale project since it doesn’t offer failover, replication, or use a database, though I know some larger scale projects that have/do used it. Though you can add some customization, you’re still locked in to a vendor which can be good or bad.

So:

  1. You should be able to find companies through a google search or individuals through a recruiter.
  2. That would very much depend on the scale of the project. I’m currently converting about 5000 phones to Asterisk and with new PRI hardware, servers, cabling, phones, and other equipment. My budget was about $600,000 which was half of proprietary vendors.
  3. Support. You’ll need someone to become an administrator of the system if you want to manage it in-house, make changes, or add features.
1 Like

Thanks for the reply and I welcome any other replies as well.

Our project is decisively smaller at onset with 5 sites and less than 20 phones, but our roadmap is one where the scaling on numbers of sites, in particular, is a factor where we see the physical sites doubling and number of phones increasing 5-6 fold within the next year.

Indeed, we understand FreePBX could be used, though the counsel received thus far is, partly because of the reasons you mentioned, for us to explore an Asterisk-based solution and consider server use through Google’s GCP so support from tech/admin management can focus on updates and overall system integrity instead of hardware.

Thanks for the recruiter idea… if has recommendations for trusted/effective recruiters or developers, please do share.

The big problem we see here with this sort of development is that someone new gets recruited to look after the system and has absolutely no idea how it works; we have to tell them that Asterisk is a tool kit and they are responsible for understanding how it is glued together, as there isn’t a standard dialplan.

Very valid point; hence, I think it most wise to select only more seasoned developers with a trackrecord of deploying projects successfully.

I can’t speak to GCP as I’ve never used it. Your spec seems mild, so the hardware would be minimal. I’m running 3 separate clinics, each on a single Intel NUC i7 (paired in failover) with over 200 extensions. If you figure $100 per phone x 120 phones, plus 2 PRIs (depending on your volume), plus hardware if you choose not to go with Google, you’re probably looking at $16,000-$20,000 USD in upfront costs, plus $800-$1200 per month for the PRIs. SIP trunks are also an option and are sometimes cheaper.

Like David said, Asterisk is a framework. By itself, it really does nothing. If you have a bright person on staff, they may be able to extrapolate all they need from an Asterisk book or by going to Digium training.

Hi @mkozusnik,
The deeper into the discovery we go is the more I concur with your initial comment. It is actually “mild”. When you mention that you currently are running three separate clinics, did you build out our clone this solution yourself on asterisks or freepbx?
If so, would you be able to build and configure one for us? And if so, at what cost? We would opt for GCP instead of buying servers, primarily to avoid the capital cost and ongoing equipment maintenance (nothing unique… same story as everyone else pushing to cloud).
Do get back to me if you are interested.

I built the environment on asterisk from source. I needed the database replication and failover capabilities. Honestly, I’m sure there are quite a few people here who could build a system, but I suspect most will be hesitant. I spent the better part of 10 months building this system and 4 web interfaces. I didn’t get much direction until I was mostly done. Luckily, I had experience building our other systems and a telecom guy who was familiar with common features, and some uncommon. Yet, I still get new feature requests. “Can you make the lunch-time queue closed message automatically be different than the after hours or no agents closed message?” That one came in on Friday. :face_with_raised_eyebrow:

There are a few things developers don’t like to tolerate; ongoing maintenance, constant modification and feature requests, and scope creep. You’ll need to define the specifics of your request before anyone would likely offer to help build it. In just a couple hours, most anyone here could build a system from scratch and have it making calls between endpoints, but if you ask for IVRs, Queues, Parking, Paging, blacklisting, forwarding, pickup groups, call recording, reporting, findme/followme, conference bridges, ring groups, phone provisioning, voicemail, etc. you’d be asking for quite a lot of time to be invested. Also, if you’re looking for 99.999% uptime, you’ll be adding time for clustering and network upgrades.

People do this out of necessity and/or the love of open source and Asterisk. I don’t think I can share the system I assembled because my hospital paid me to build most of it, but perhaps someone else here has a pre-built system they would be willing to share or sell.

1 Like

@mkozusnik

Perhaps you can inquire what they might want as a licensing cost for selling an instance of it?

Even if that were to happen, you’d still be in the same situation. You have nobody to manage the system at this time. This is the user interface I built for the Admin tools. Something like FreePBX has most of these built in also (and it’s free). The problem would be that someone on your side would need to understand what the options mean and how they interact with the system. There would be a learning curve equal to, if not moreso than a package.

@mkozusnik

We have a small IT dev team which I sense ought to be able to handle ongoing management of this tool. Grant it, I could be wrong, but I genuinely get the impression that since they are literally building custom tools in-house, this might eventually be within their wheelhouse to handle.

Aside from the 18 servers/VMs I’m running, I also have a central MySQL database cluster, about 30 PHP programs on the servers, and a cluster of apache web servers. I suspect that this system would be overkill for what you’re proposing. If I were to build it on a small scale, it may still take a month or two to build in all of the features. Also, we chose 3 models of phones and the provisioning only creates the config files for those models.

What features/functions are you looking to implement? Meaning, what is the system expected to do?

@mkozusnik

I can forward you an outline of core needs versus nice-to-haves. Before moving forward I’d like to get a ballpark budget of what you anticipate. I’m sure with your level of expertise you can assess this and if it is within range then it makes sense to continue. If not, then I don’t want to waste your time.

That’s fine. I’d be happy to look at the spec.You can paste it here or PM or I can give you an email.

Please advise which email should be used.

mwkoz3@comcast.net … I don’t watch it closely, but I’ll keep an eye out.