Can anyone help me with this error. I would like to know what this error means. From what i can see its a call disconnect but it a fault or maybe a call restriction?
%VOICE_IEC-3-GW: H323: Internal Error (Socket error): IEC=18.104.22.168.7.5 on callID 6732 GUID=4DD81BA230AA11E385F070CA9BB22E20.
CME-2921#show voice iec description 22.214.171.124.7.5
IEC Version: 1
Entity: 1 (Gateway)
Category: 186 (Signaling socket failure)
Subsystem: 5 (H323)
Error: 7 (Socket error)
Diagnostic Code: 5
We will need further investigation.
Can you please post the complete "sh run" and "sh version"?
We will also need to know the complete call flow, so we can suggest what debugs to capture for further troubleshooting.
-- Jorge Armijo Please remember to rate helpful responses and identify helpful or correct answers.
I could reproduce something in the lab which causes the same error codes:
287674: Oct 14 14:20:30.840: ISDN Se0/0/0:15 Q931: TX -> CALL_PROC pd = 8 callref = 0xFA49
Channel ID i = 0xA98388
Exclusive, Channel 8
287675: Oct 14 14:20:36.840: %VOICE_IEC-3-GW: H323: Internal Error (Socket error): IEC=126.96.36.199.7.5 on callID 272232 GUID=D9FEA303340111E38068001DA10435C0
287676: Oct 14 14:20:36.848: ISDN Se0/0/0:15 Q931: TX -> DISCONNECT pd = 8 callref = 0xFA49
Cause i = 0x80A6 - Network out of order
This happens in my case if an interface has the H.323 bind whose IP address is not reachable from the CUCM (loopback interface, routing issue?).
No TCP connection can be established between gateway and CUCM.
A good debug choice on the gateway is "debug ip tcp transaction"
Here my example:
287691: Oct 14 14:21:58.222: TCP: Random local port generated 27825, network 1
287692: Oct 14 14:21:58.222: TCB4B0EED38 bound to 10.0.0.1.27825 <--- IS THIS THE INTERFACE YOU WANT FOR SIGNALLING??
287693: Oct 14 14:21:58.222: Reserved port 27825 in Transport Port Agent for TCP IP type 1
287694: Oct 14 14:21:58.222: TCP: sending SYN, seq 227953095, ack 0
287695: Oct 14 14:21:58.226: TCP0: Connection to 10.45.0.50:1720, advertising MSS 1460
287696: Oct 14 14:21:58.226: TCP0: state was CLOSED -> SYNSENT [27825 -> 10.45.0.50(1720)]
287697: Oct 14 14:21:58.226: ISDN Se0/0/0:15 Q931: TX -> CALL_PROC pd = 8 callref = 0xFB49
Channel ID i = 0xA98389
Exclusive, Channel 9
287698: Oct 14 14:22:00.226: 10.0.0.1:27825 <---> 10.45.0.50:1720 congestion window changes
287699: Oct 14 14:22:00.226: cwnd from 1460 to 1460, ssthresh from 65535 to 2920
287700: Oct 14 14:22:00.226: TCP0: timeout #1 - timeout is 4000 ms, seq 227953095
287701: Oct 14 14:22:00.226: TCP: (27825) -> 10.45.0.50(1720)
287702: Oct 14 14:22:02.634: TCB4B0A7300 setting property TCP_TOS (11) 498A5BD8
287703: Oct 14 14:22:02.838: TCP0: ACK timeout timer expired <--- NO ACK RECEIVED FROM CUCM
Check on the CUCM with CLI "utils network capture host ip " if packets from the gateway actually arrive at CUCM. If they do maybe the way back is not possible.
Try to telnet the CUCM to port 1720 from the gateway signalling interface.
If you don't get an "open" you have a network issue (routing, firewall, ACL...)
router#telnet 10.45.0.50 1720 /source-interface lo10
Trying 10.45.0.50, 1720 ...
% Connection timed out; remote host not responding
router#telnet 10.45.0.50 1720 /source-interface fas0/0
Trying 10.45.0.50, 1720 ... Open
hope that helps
In this case check the IP Address which is showing in CUCM for H.323 gateway configuration and the interface IP Address with which you have bind your H.323 gateway.
This happens due to mismatch in both IP Address i.e CUCM side and router side of H.323 gateway.