Symptom
When we execute the ping command on the cli from any of the devices to reach a destination inside the vpn 110 it is working:
command: ping vrf 110
result: destination-fqdn is resolved by dns server wich is reachable via service vpn 110.
When we execute the ssh command on the cli from any of the devices to reach a destination inside the vpn 110 it is NOT working:
command: ssh -vrf 110
result: the destination-fqdn was not resolved and theirfore, the ssh session could not be established.
Conditions
Devices using 17.4.1b
ip name-server vrf 10 x.x.x.x x.x.x.x
ip domain lookup vrf 10 source-interface Loopback10
Further Problem Description
Confirmed that when SSH command is used device use vpn 0 as source and not the proper vpn (vrf) this cause the failure.
When we use the SSH request, device use source the vpn 0 interfaces and as there is not name server for DNS on global configuration (only dns configuration for the service side vrf) send the dns request to the broadcast. So here the problem is that when the ssh is done the device don’t use VRF 110 routing information or configuration at all.
no ip ssh source command available to force this to use proper source and routing table.
issue similar to CSCvw82778
