Topology:
SIP Circuite -> Austin CUBE -> DATA CLOUD -> Houstone Gateway -> CUCM
Issue:
Intermittently when placing SIP Calls on hold in Austin the call is lost due to CUBE throwing an Internal Error.
Cause: This is due to CUBE Receiving from CUCM the next Re-Invite before receiving the ACK from CUCM.
PCAP traces on the Houston side confirm the messages are sent from CUCM in the correct order, ACK before next Re-Invite.
However, Austin CUBE pcap shows the ACK is received after the next re-invite.
I believe this is caused by a race condition within the WAN, and rarely the Re-invite make it to Austin before the ACK.
Question:
Is there anything I can do from CUBE or CUCM perspective to combat this rare condition?
I have a ticket open with the WAN provide, but not sure if anything can be done.