I have a QPID daemon running inside of a Docker container. I have C++ code that should connect to this daemon, and in the past, it was able to, but it suddenly stopped being able to connect when the C++ code was run on the host machine.
Trying to connect to the daemon on the container from the host throws a qpid::messaging::TransportFailure expection. When I launch qpidd with -t for trace level logging, I see the following each time a connection is attempted: 2017-07-12 15:59:33 debug RECV [172.17.0.2:5672-172.17.0.1:37696]: INIT(0-0) 2017-07-12 15:59:33 debug SENT [172.17.0.2:5672-172.17.0.1:37696]: INIT(0-10) I would guess that this has something to do with the ActiveMQ protocol, but I am not sure. I don't think the fault lies with the code, because the exact same code running on the same container as the broker is able to connect to the QPID broker. I know that the daemon is visible from the host machine because curl finds the daemon Why can I not connect to the daemon from the host, even though I can from the container that the daemon is running in? It may be worth noting that the versions of QPID on the host are different than the container. The host has the latest version on CentOS 7 (1.36.0) while the container has the latest version on Ubuntu 16.04 (0.16). Thanks Matt
