Modem as a service?

Larry Sheldon larrysheldon at
Mon Dec 7 17:54:17 UTC 2015

On 12/6/2015 16:17, Karl Auer wrote:
> On Sun, 2015-12-06 at 16:36 -0500, James R Cutler wrote:
>>> On Dec 6, 2015, at 2:19 PM, James Laszko <jamesl at> wrote:
>>> ... we don’t need to actually connect to the OOB modem on the other side, we just need a NO ANSWER/ANSWER kind of response. …
>> Forget modems - to probe via some kind of analog connection, just get
>> a single instrument wireless telephone with answering capability.  For
>> a bonus, put some kind of identifier in the answering message:  No
>> power > no answer; power > answer.
> I must be thick - how does that solve the problem? The OP wants to know
> if a modem at a remote site will answer the phone. Maybe I misunderstood
> the problem.

I'll join the confusion--I thought the OP wanted to test for power 
availability at the distant site by seeing if a modem there would answer 
the phone there.  That it HAD to be a modem in that case makes no sense 
to me.

I'm of the line now and have been for a while and maybe y'all don't do 
things the way we did--we always had an answering machine (two or three 
in some places*) that always answered on the first ring and gave some 
kind of status report that was updated hourly on on event).  If it did 
not answer, the power was out.

*at one site we had one that gave general status--what's up, what's 
down, what's generally interesting (outages scheduled soon, where we are 
in the daily batch cycle).  We had another listing southern region 
outputs ready for pick-up and one listing northern region stuff.

sed quis custodiet ipsos custodes? (Juvenal)

More information about the NANOG mailing list