[CM305004]: Failed to find a transport to host

Discussion in '3CX Phone System - General' started by matthijsolislagers, Jan 22, 2008.

  1. matthijsolislagers

    matthijsolislagers New Member

    Joined:
    Jan 22, 2008
    Messages:
    140
    Likes Received:
    0
    Hi,

    I'm running 3CX build 5.0.3790.0 on a Windows server 2003 SBS. The server is multihomed with one external nic and one internal nic. The external nic has a external ip.

    When I'm trying to register with XeloQ I get the following message in my server log:

    14:45:59.017 DeviceRecord::DeviceRecord [CM305004]: Failed to find a transport to host 81.26.212.150

    I tried to remove the stun server but it didn't work.

    The server log show's both IP's:

    14:39:55.249 CallMgr::eek:nAddIPs IP(s) added:[10.0.0.1,xxx.xxx.31.238]
    14:39:45.921 CallMgr::findLocalIPs [CM501006]: Default Local IP address: [10.0.0.1, xxx.xxx.31.238]

    Can someone please help me? I'm stuck..

    I'm planning to buy the Small business version when this works.
     
  2. matthijsolislagers

    matthijsolislagers New Member

    Joined:
    Jan 22, 2008
    Messages:
    140
    Likes Received:
    0
    OK. After a lot of reboots, sleepless nights, reading logs I figured something out.

    I noticed that after a random time the line does register successful. After a reboot the same issue began.

    Here's wat I've figured out:

    1) The server has 2 nic's: First: Internal (ip: 10.0.0.1) Second: External (ip: xx.xx.31.238)
    2) The server is also a VPN server (routing and remote access)
    3) When I reboot the server the log is as following:

    15:30:47.027 DeviceRecord::DeviceRecord [CM305004]: Failed to find a transport to host 194.221.62.198
    15:30:46.996 IVRConnected [CM501004]: Connected: IVR Service
    15:30:46.980 MediaServerConnected [CM112000] Media Server is connected
    15:30:46.855 DBA [CM501003]: Connected: Database
    15:30:46.777 CallMgr::findLocalIPs [CM501006]: Default Local IP address: [10.0.0.1, xx.xxx.31.238]
    15:30:46.621 CallCtrl::thread [CM501007]: *** Started Calls Controller thread ***
    15:30:46.621 CallMgr::Initialize [CM501002]: Version: 5.0.3790.0
    15:30:46.621 CallMgr::Initialize [CM501001]: Start 3CX PhoneSystem Call Manager
    15:30:46.621 LoadLicenceInfo [CM501010]: License Info: Load Failed

    4) When a random home worker connects a VPN tunnel to the server the following log entry's show up:

    15:34:09.648 ClientRegs::eek:nSuccess [CM504004]: Registration succeeded for: 10000@Voipbuster
    15:34:09.289 ClientRegs::Register [CM504003]: Sent registration request for 10000@Voipbuster
    15:34:08.023 StunClient::process [CM506002]: Resolved SIP external IP:port (xx.xxx.31.238:64285) on Transport 10.0.0.106:5060
    15:34:07.851 StunClient::eek:nAddTransport [CM506001]: STUN request to resolve SIP external IP:port mapping is sent to STUN server 64.69.76.23:3478 over Transport 10.0.0.106:5060
    15:34:07.726 CallMgr::eek:nAddIPs IP(s) added:[10.0.0.106]

    5) So, it seems the random time is actually not random. It is a random VPN worker.

    Now the strange thing is.. Transport 10.0.0.106 is actually a virtual ip. Created by routing and remote access for VPN purposes. There is no gateway or internet connectivity possible via that ip. So to me it seems like a bug in 3CX. The only way to the internet is through the external nic. 3CX doesn't try that.

    Can anyone confirm this? Has someone got a solution for this?

    Thank you!
     
  3. matthijsolislagers

    matthijsolislagers New Member

    Joined:
    Jan 22, 2008
    Messages:
    140
    Likes Received:
    0
    Maybe a feature in 3CX to select the external nic is a solution for this problem. I notices more users having the same problem.
     
  4. matthijsolislagers

    matthijsolislagers New Member

    Joined:
    Jan 22, 2008
    Messages:
    140
    Likes Received:
    0
    Does someone have the same problem as I do?
     
  5. Nick Galea

    Nick Galea Site Admin

    Joined:
    Jun 6, 2006
    Messages:
    1,732
    Likes Received:
    42
    So reading your topic - a new connection by a VPN worker registers a new IP on the system (that of the remote VPN worker) and this causes registrations to go via the VPN connection which fails?
     
  6. matthijsolislagers

    matthijsolislagers New Member

    Joined:
    Jan 22, 2008
    Messages:
    140
    Likes Received:
    0
    No, not really.

    Before the VPN worker log's in the 3CX server notices 2 IP's as you can see in the log: 1 internal and 1 external. At this point 3CX is unable to register with the VOIP providers and is unable to contact STUN.

    After a VPN log's in another IP is noticed by 3CX. The local endpoint IP of the VPN (the gateway IP for the VPN home user). How this exactly works I don't know, but this is the way Routing and Remote Access works. I don't think the registration goes via the VPN tunnel. I can see the connection go via the external IP when I monitor it. 3CX just thinks it does..

    So my problem is. After a reboot 3CX doesn't register with the VOIP providers until a VPN home worker connects to the server.

    I hope this is clear enough for you. If you need any more information just let me know.

    Thanks
     
  7. matthijsolislagers

    matthijsolislagers New Member

    Joined:
    Jan 22, 2008
    Messages:
    140
    Likes Received:
    0
    For your information, this is my routing table:

    Image no longer available
     
  8. SY

    SY Well-Known Member
    3CX Support

    Joined:
    Jan 26, 2007
    Messages:
    1,821
    Likes Received:
    1
    matthijsolislagers,

    Please find [Network] section in 3CXPhoneSystem.ini file and add value(or modify it if exists) localSubnets=0.0.0.0/0

    [Network]
    localSubnets=0.0.0.0/0
    ....

    then restart PBX.

    It should resolve this problems.
    Please send me confirmation(or confutation) to private, or "stepan" email box.

    Thanks
     

Share This Page