One user with two phones and two 911 locations


#1

I’m looking for some suggestions on how you might handle this issue.

I’ve got several users now who have a phone at home and at the office. I use pjsip, so I could just have both phones on the same extension… but then how do I assign a different 911 caller ID for each?

For now, i’ve created new extensions and turned on follow me so that both phones ring when someone calls the main extension. But then the BLF key for that user doesn’t show as busy when they’re on a call at home. Perhaps there’s some easy way to have a BLF key monitor more than one extension and that’s the best solution?

Thanks for any thoughts!


(Matthew Fredrickson) #2

As I recall, if you’re using one PJSIP extension with two contacts per AOR (basically two phones registering with the same credentials) which it sounds like you’re doing. I don’t think there’s a way to have separate 911 Caller IDs in that case.

@jphilip or @wmoon might correct me if I’m wrong though.


(Philip Joseph) #3

Hi , @mattf
You are right, Since there is only one user you can not set different E 911 caller-id.
Only for HOT desking we have this feature , location wise E911 callerid

Regards
Philip


#4

Thanks for the quick replies!

I’ve found this older thread with instructions on modifying the hints so that the main BLF key will show in use when either phone is in use: BLF to monitor multiple extensions

That seems to be doing the trick.


(Jared Busch) #5

Remote worker 911 location is also not yet a requirement, only a nice to have. It will be a requirement, just not yet.

Right now, your policy for remote workers should be clearly stated, maybe even get a (digital) sign-off from each employee, that they know and understand that 911 does not work correctly from their phone at home.


(Jared Busch) #6

It is more than 911 that does not handle multiple AORs correctly.


(Itzik) #7

So I see I’m not the only one that uses PJSIP across locations…

I politely ask the FreePBX team to rethink the implementation of setting e911 CID based on IP.


(Speedy01) #8

@TechGuyMike
This wont help your home users, this is how I’m handling e911 for the same extension over multiple locations:

@PitzKey : Maybe the above will be helpful for you as well.


(Tom Ray) #9

This still doesn’t solve anything. 1) Remote phones could be at a location that requires multiple 911 addresses. They could have a first floor and a second floor phone, they both have the same IP to the remote PBX (their WAN). 2) Dynamic IPs for remote locations. Today my IP could be X and tomorrow my IP could be Y. In three hours my IP could be Z. That will depend on different factors.

Many devices allow their MACs to be sent in headers. There is also the fact that the From Name in devices isn’t used by FreePBX at all. I have places where I have 8 port FXS gateways for their FXS lines and I use the From Name setting to set Line 1, Line 2, Line 3 and I use that to determine which line is being used to make the call.

IPs can change, they are not always static. Some devices are behind crappy routers that do poor NAT handling so the location IP you see in the contacts isn’t the Received IP being used to send traffic to. That could mess with things Basing things off IPs would mean you knew every IP being used and that’s not possible in all setups.


(Jared Busch) #10

If by IP you assume he means by AOR, this can easily solve the problem for many implementations.

Sure it is not all use cases, but it is many.

Also until PSAP can accept more than a damned DID and then carriers update their system, there are no good solutions to the new 911 location rulings anyway.

This is my extension on my PBX right now.

    Contact:  103/sip:103@12.XXX.XXX.XXX:1030               5689f9ce21 Avail        17.223
    Contact:  103/sip:103@64.XXX.XXX.XXX:12467;transport=TLS b4268755d4 Avail        23.501
    Contact:  103/sip:103@38.XXX.XXX.XXX:1130                eb2ddaa931 Avail        10.669

Two of those three AOR are on known static IP.
So if the outbound emergency CID logic looked at the AOR, it is easy to write comparison logic to choose an emergency CID based on “matches” or “begins with” or some such.


(Tom Ray) #11

So let’s take your example and make this change:

So now I have an extenstion at the same location on two different phones in to different physical locations requiring two 911 locations. Now I could look at the full Contact URI and match up to that but that would require that Contact URI to be the same always. That’s not going to happen.

How do I then determine which phone is making that call? Is it the one on the first floor or the one on the second floor? Of course this is assuming they have a static WAN IP and I just have to deal with the changing ports. If they have a dynamic IP then that WAN IP could change just like the port could.

So as I pointed out, if the device supports it you could enable sending the MAC as a SIP header. You could also use the From Name of the device to determine which device is sending the call.

I just don’t think an IP based solution is the right way to do this. There are to many factors which make it too much of a moving target.


(Pakenvs) #12

What we do is:

We use this for hot-desking and multiple contacts with pjsip so phones have the same extension.
Modify the dialplan on the phones themselves to prepend a “site code” to 911, then catch that code in an outbound route that delivers the proper 911 caller id.

As an example in our Grandstream phones we add this to the dial plan: <=30>911

then in our outbound route, we catch 30911 and send it to the 911 provider with the proper caller id.

This was the cleanest way we could find for multiple phones for single user and 911 management.


(system) closed #13

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.