Vodia PBX

This is what get: # host -t NAPTR ssw3.brussels.weepee.org # host -t SRV _sips._tcp.ssw3.brussels.weepee.org Host _sips._tcp.ssw3.brussels.weepee.org not found: 3(NXDOMAIN) # host -t SRV _sip._tcp.ssw3.brussels.weepee.org Host _sip._tcp.ssw3.brussels.weepee.org not found: 3(NXDOMAIN) # host -t SRV _sip._udp.ssw3.brussels.weepee.org Host _sip._udp.ssw3.brussels.weepee.org not found: 3(NXDOMAIN) # host -t AAAA ssw3.brussels.weepee.org # host -t A ssw3.brussels.weepee.org ssw3.brussels.weepee.org has address 91.208.12.133 The record expires in approx. one hour, that all looks fine. The only problem that I would see with DNS is that you are using the router DNS server which does not support NAPTR or SRV or AAAA records, we have seen cases where this screwed up the DNS lookup. However because the first lookup is okay, this is probably not the problem. You should also check if the router has some SIP awareness which might also screw it up. if it all does not help, my recommendation would be to install a tool like Wireshark and run it when the PBX fails. You should see the DNS queries, the SIP queries and that should then help us to find out where the problem is.

Oh sorry we are talking about the PBX... The PBX mixes the SIP messages in the log, while the phone has it seperated (which makes it difficult to match it to log messages).

Thats all good and nice, but I dont see any messages from the PBX... Your logging settings should be log level 9, enable everything and you can actually turn SIP messages off now. Sorry for this ping pong, but we need to see the part where it runs the call through the dial plan.

What is your outbound proxy? Is it a DNS name? If yes, maybe you can tell us and we can take a look if there is something suspicious there.

That MESSAGE is not really the point. Do you have the log with messages at log level 9? To make sure you are not missing the "Evaluation" log message, make a outbound call where it should go outbound. If you can see the Evaluating message, then make the call to the number above and then we should be able to see where the problem is.

Remote without touching the phone requires that you talk to snom and they set up the redirection service for you MAC addresses.

Well the problem could be that the call connects as soon as you dial the number, then the hunt group would obviously stop. For example, when calling the cell phone the mailbox could pick up. Terminating traffic in the analog world is not so wasy and many gateways send the connected signal immediatly (you will still just hear ringback tone); that's because it is so hard to figure out if the call is connected or not in the analog world. You can check in the SIP messages if the gateway sends a 200 Ok response on the INVITE request.

You could try to narrow it down by registering a softphone and see if that works. Maybe it is something simple like the Internet is down or the service provider is not available (e.g. under DoS attack).

Yea, the phone is a client there things are pretty simple... Plus there are typically more (dumb) phone installations than PBX installations.

This will be handled by snom now. The persons are the same, and there are more support people in snom available who can also take over.

Yes. If you want to talk to the PBX, you need a routable address and the PBX must be able to advertize this address to the phones. Sounds trivial, but it's the core of the problem.

Well, NAT is probably the biggest problem with VoIP. Until IPv6 is ready, we need to deal with more or less dirty workarounds. Check out http://kiwi.pbxnsip.com/index.php/Office_with_private_and_public_IP_addresses for a typical scenario. If you want to register phones from the Internet, you do need to have a routable IP address ("public" IP address); all the workarounds with port forwarding etc are extremly difficult and instable so consider putting one interface of the PBX host on a public IP.

I can understand that frustration well . It is not only VoIP also other products are getting so complex you need a PhD to get them working (cell phones, cars, home automation to to name a few).

That is a Mac-specific problem; it is not really a reason for concern unless you really want to use TOS to give voice packets a higher routing priority.

Here is another version, which automatically picks another port if port 80 should not be available. It also shows if the phone is registered or not.

Not yet, there will be another one coming out (Q/A found another problem).

pbxnsipsupport will be moved to the wiki and the jira (I believe). IMHO a good move as the document system on pbxnsipsupport was not world class and jira has the advantage that there is one central trouble ticket system for all products.

Forget about the m9 soft phone so far. Maybe the latest PBX software will solve your problem already. If not, I would try to run the X-lite in the same LAN segment as the PBX. If it still does not work, there must be something wrong in the dialplan area.

Make sure you don't use STUN on X-lite, which is a permanent (and unneccessary) source for trouble. Also, we are coming out with a updated version of the PBX this week, if you can backup and update your PBX and see if that fixes any issues: Win32 – http://pbxnsip.com/download/pbxctrl-2011-4.2.0.3974.exe Win64 – http://pbxnsip.com/download/pbxctrl-2011-4.2.0.3974_64bit.exe Debian - http://pbxnsip.com/download/pbxctrl-debian4.0-2011-4.2.0.3974 Centos32 - http://pbxnsip.com/download/pbxctrl-centos5-2011-4.2.0.3974 Centos64 - http://pbxnsip.com/download/pbxctrl-centos5-2011-4.2.0.3974_64bit SuSe - http://pbxnsip.com/download/pbxctrl-suse10-2011-4.2.0.3974 Sheeva – http://pbxnsip.com/download/pbxctrl-sheeva-2011-4.2.0.3974 You can also try the snom m9 soft phone, which should work without trouble (http://forum.pbxnsip.com/index.php?/topic/4140-snom-m9-soft-phone/).

Well, thats a SIP problem. How do you successfully disconnect a call that is not connected yet? We could send 487, right now we are sending 486. Not sure if this would make a big difference.

# host -t NAPTR sipgate.co.uk # host -t SRV _sips._tcp.sipgate.co.uk Host _sips._tcp.sipgate.co.uk not found: 3(NXDOMAIN) # host -t SRV _sip._tcp.sipgate.co.uk Host _sip._tcp.sipgate.co.uk not found: 3(NXDOMAIN) # host -t SRV _sip._udp.sipgate.co.uk _sip._udp.sipgate.co.uk has SRV record 0 0 5060 sipgate.co.uk. # host -t AAAA sipgate.co.uk # host -t A sipgate.co.uk sipgate.co.uk has address 217.10.79.23 That is what the PBX should resolve.

Hmmm. As you dial plan does not contain the entry "dial extension", it must have to do with something else than the dial plan. Does the phone send the right number? Maybe there is a local dial plan on the phone which does the magic. Also, what does the log of the PBX say? On log level 9, there must be something like "Dialplan: Evaluating xxx against xxx".

CentOS is like RedHat. Can you start the PBX manually? Try starting it with pbxctrl --dir /bla/dir --no-daemon --log 9. Then you see if for example there is another service running on port 80.

Check out these links: http://wiki.snomone.com/index.php?title=Quick_Installation#Linux_Installations and the old one http://kiwi.pbxnsip.com/index.php/Installing_in_Linux

What version are you on? Does it work with no cell phones involved (just regular extensions with no cell phone forwarding)?