A peering connection enables connectivity between two VPCs, but the connectivity cannot be transited.
For example, peering connections are created between VPC 1 and VPC 2, and between VPC 1 and VPC 3. However, due to the non-transitivity of peering connections, traffic cannot flow between VPC 2 and VPC 3.
Even if a peering connection is created, both ends of the connection cannot communicate with each other if routes for sending and returning packets are not configured at both ends.
For more information, see Features.
No. The peering connection can be disconnected at either end at any time. The connection is ineffective immediately upon disconnection. The VPC at either end is accessible only after the connection is created again.
Yes. This is possible as long as the owner of the other VPC accepts your peering connection request.
No. The IP ranges of VPCs at the two ends of a peering connection cannot overlap.
No. After a peering connection is created, access between two VPCs is similar to that between two CVMs in the same VPC, without additional encryption. Network traffic within a VPC is always isolated from other networks.
Single points of failure or bandwidth bottlenecks cannot occur because a peering connection between VPCs is not a gateway or a VPN connection and it does not depend on separate hardware.