Wired access to SMS?

Seth Mattinen sethm at rollernet.us
Wed Oct 10 17:36:33 UTC 2012


On 10/10/12 10:10 AM, William Herrin wrote:
> On Wed, Oct 10, 2012 at 12:18 PM, jamie rishaw <j at arpa.com> wrote:
>> On Tue, Oct 9, 2012 at 7:15 PM, Aaron Toponce
>>> Instead, purchase a cellular USB modem with a standard plan. All 4 major
>>> carriers provide APIs to interact with the modems, and you get everything
>>> you need*. They aren't cheap (something in the neighborhood of $30/month),
> 
>>     If his need is mission critical, and $30/mo breaks the bank .. I'd
>> respectfully submit that there wasn't much of a mission.. :-p
>>
>>     I do agree, tho, that an external / serial / mmmmaybe-usb gsm device is
>> the route to pursue.
> 
> Perhaps I should explain a little further:
> 
> I have a system in place based on just under a dozen Multitech GSM
> modems in a room by a window. It works... more or less.
> 
> It has no provisions for equipment or site failure. The modem breaks,
> that number is unavailable. The site fails, that number is
> unavailable. The local cell network gets jammed, the number is
> unavailable. That's the opposite of "high availability."
> 
> So, I need to replace it with something that offers high availability
> for each phone number, aka "SMS long code." I realize that the phone
> end will still suffer all the vagaries of SMS. But on the base end I
> need high availability.
> 
> I expect this to cost more than throwing a dozen GSM modems in a room.
> I won't be offended when it does.
> 

What about finding someplace offsite and setting up a persistent PPP
connection with modems (of the POTS variety) between it and home base?
Put half the modems there and maybe a low power Atom server with hooks
to send alerts like "connection to home hasn't come back after X redials".

I do something similar by having cheap DSL with a provider I don't have
any other services with to provide a outside world view of things. I
have a POTS line there too that can auto-dial back home if needed.

~Seth




More information about the NANOG mailing list