When I "write mem" and reload the router, I can't seem to log in again. I get a "password" prompt instead of the SSH key authorization being accepted.
There seems to be an issue with adding additional interfaces to the 1000V. If an interface is added and associated prior to the bringup of the instance, the interface shows up as DHCP and the router's management becomes nonresponsive. If an interface is added after the instance is up everything seems to be OK (interface shows as present, no address, and shutdown). If the interface is given an address, it can't seem to ping the other instances on a given subnet in the same availability zone and VPC. Yes, I have confirmed my security group, network ACL, interface ACL, and route table configuration.
Sometimes when the CSR1000V is reloaded from the AWS console, it will not boot up at all, showing strange messages in the log about no compatible software. This seems to be mitigated if the instance is "stopped" rather than "rebooted".
A common use case is the ability to use the CSR1000V as a lab router. In some of these use cases it is desired to have the CSR1000V start up inside of a "private" subnet within a VPC with no internet gateway or external routing attached. By doing this, you can't SSH into the instance.
Can anyone suggest any workarounds for the above, or the correct procedure to have a CSR1000V with multiple interfaces?
The following document contains examples of how to deploy and configure
the CSR1000V within Amazon Web Services (AWS) in a DMVPN configuration
as well as details on how to setup VPC Gateway Redundancy.