Vodia PBX

We need to check this.

That really sounds not like it should be. The part where the caller calls in an hears MoH I dont get it. Does the phone accept the call and put it on hold right away? What phones are you using?

At the end of the day, the PBX can send a report which shows how active an agent was. So if the issue is lazy agents, it should be possible to see how long and how much an agent worked. The number of missed calls per agent is missing though right now, good point. We could include that in the statistics.

At 08:28:30 (15 seconds after switching to T.38) the "MG4000|2.0" wants to get back to regular voice. At least it sends a Re-INVITE without T.38. I guess that is the problem here. Maybe it does not like the T.38. Is there anything in the log of this device? That should should contain the hint why this call fails. From the PBX perspective it is hard to say. Until the Re-INVITE the call looks normal.

So far people seem to use an ATA for door intercom. Also interesting is a snom PA1, which is essentially a snom 300 phone in a different housing. Both should be working as good and as bad as the desktop versions.

Usually when you are in the US, you can solve these kind of problems easily by choosing the country code "1", then the PBX will automatically put the number into the usual form and the user should see the 10-digit number on the display and is also able to call this number back.

Well, Win64 is something unusual, still these days. IMHO it should not make a difference, but maybe there is really something different. If you can give it a try on a 32-bit system and it behaves different, I would say that would be worth a try.

Yea you are right. However we are just renting this forum (hosted services!) and have no idea to make such changes...

Send a PM to pbx_support for a build. We dont want to rush it, we are still finding issues and want to make sure that the upgrade is a positive experience!

The CS410 is not a router, to answer that part of the question. So the computer will not be able to get "online"... However, the CS410 should be able to pull a NTP timestamp and if you are using the build-in NTP server of the CS410, then the phones in the LAN should also get a time. You can also call from the LAN into the internet.

We already added that in head (4.0), so it will be available eventually. White space does not look very good, but probably the most pragmatic way right now.

Anyway, ":57948" means that the PBX did not bind it to any ports. That is very strange. I cant imagine a situation where this can happen. The only thing I can think of would be that the host does not have a IP address at all (maybe temporarily because the link got lost or the DHCP server did not renew the lease).

I guess in 3.2, you have to assign the ANI to the trunk.

The PBX focus on those fields: From, To (they are taken literally). The other field is the local party, usually identified by the Request-URI or the To-header, this can be set up by the trunk inbound routing patterns. This is the one who has to pay for the call. Then the PBX also reports the remote party, which may differ from the To/From header, e.g. in the case of redirection.

In the ANI field, you can actually specify a list of ANI, seperated by space. If you put the trunk name in front of it (followed by a colon), then the ANI sticks to that trunk. Would that help? E.g. "9787462777 Trunk1:21212345". This implies that trunk names must not use space characters if you want this to work.

Well, the CS410 does not have a battery-powered realtime clock. If you cant get to the Internet, I see only two solutions. First, use a PC-based PBX host that has all this stuff and you can just set the clock during the installation. Second is to install to a NTP server on a PC in the office. If you are able to run a NTP service on that computer, you can probably as well run the PBX service there...

That worries me. It should say something like "Opening socket on 0.0.0.0:57948". What is the content of "port_bind4" in the pbx.xml directory? Maybe there is some garbage that tells the PBX to bind the ports to something which is impossible.

You mean the phone ring melody should be different? You can't influence that for calls directly to your extension, but you can when the call goes through a hunt group or a ACD. Otherwise if you want to have two different phones ring, you need to use different extensions.

The names in the ringtones.xml file must have translations. These translations are in a global dictionary, which you probably don't want to change. I thing we must add something so that the blanks are filled with a name from the ringtones.xml file, even if there is no translation available.

Well, the PBX has to decode both media streams before they are mixed. Then there is another transcoding going on, because usually the PBX saves the recording in GSM format.

I guess the PBX can only put into the CDR what it gets (if we are talking about inbound calls).

The reason is that in hosted environments the domain admins (customers) could steal DID numbers from other domains. But AFAIR you can control that in the global settings.

Can you see the REGISTER packets in the log of the PBX (turn SIP logging and and enable REGISTER messages; set log level to 7). If yes than the Firewall sends the packets to the PBX, step #1 is okay. Then maybe check the Windows firewall for it's policy about letting the PBX access the network. Maybe it blocks outgoing UDP traffic. If you have a phone that has a logging feature for the SIP traffic, you can see if the 200 Ok to the REGISTER actually makes it to the phone.

No, RPI actually requires that the "From" goes into the RPI and then the from header takes the content of the authenticated party. Sounds complicated and troublesome? That is why if never became RFC and should not be used any more. IMHO ICID is the best. 2nd best are PAI or PPI. Unfortunately there is a lot of legacy stuff out there and voting here would not help...

Did you check the article https://pbxnsipsupport.com/index.php?_m=kno...kbarticleid=651 that describes how inbound call are sent to extensions? That might be useful here.