Description of problem: When adding new Middleware Provider by specifying it's IPv6 address and trying to validate the credentials, it fails with message: "Credential validation was not successful: Unable to connect to 2620:52:0:1017:21a:4aff:fe10:1767:8080" Version-Release number of selected component (if applicable): 5.8.0.4.20170307183144_745a3d9 How reproducible: always Steps to Reproduce: 1. Configure and Run Cassandra on IPv6 address. 2. Configure and run Hawkular Services on IPv6 address and connected to C* 3. Verify that Hawkular Services is running on IPv6 address and accepts connections on 8080 port. Checked connection by nmap, curl or telnet tools from CFME machine. 4. Add new Middleware Provider by specifying IPv6 address of Hawkular Services, verify the credentials. It fails with error message: "Credential validation was not successful: Unable to connect to 2620:52:0:1017:21a:4aff:fe10:1767:8080" Actual results: Fails to connect to Hawkular Services by IPv6 address. Expected results: It should be able to connect via IPv6 address. Additional info: When specifying hostname of Hawkular Services instead of IPv6 address it works.
Created attachment 1262458 [details] validate provider
Created attachment 1262459 [details] hawkular services running
Can you retry by putting the IP address into [] as in [::1] ? Otherwise the parser can't distinguish between IP-address component and port in 2620:52:0:1017:21a:4aff:fe10:1767:8080 as the separator between address components are also colons. http://[2620:52:0:1017:21a:4aff:fe10:1767]:8080 is valid http://2620:52:0:1017:21a:4aff:fe10:1767:8080 is invalid
It did not work in either ways. Unable to connect to [2620:52:0:1017:21a:4aff:fe10:1767]:8080
Tested adding an provider both using "::1", "[::1]" with a Hawkular exposed on ipv6 only, and it worked correctly, on latest upstream master. Can you check again if that's still the case?
In fact I am testing MiQ and HawkularServices on different machines, so using IPv6 address [2620:52:0:1017:21a:4aff:fe10:1775]. Probably that is the reason that I still get the issue. Can you try to reproduce on separate machines?
Thank you Caina for your work. The problem was in network configuration, closing this BZ.