Fix error message when providing an incorrect peer-id

[chevdor]

I was using docker-compose files with old peer-id and got the following errors:

dave            | 2020-04-21 13:44:43 💔 Invalid peer ID from bootnode, expected `QmQZ8TjTqeDj3ciwr93EJ95hxfDsb9pEYDizUAbWpigtQN` at address `/ip4/172.28.1.1/tcp/30333`.
bob             | 2020-04-21 13:44:43 💔 Invalid peer ID from bootnode, expected `QmQZ8TjTqeDj3ciwr93EJ95hxfDsb9pEYDizUAbWpigtQN` at address `/ip4/172.28.1.1/tcp/30333`.
eve             | 2020-04-21 13:44:43 💔 Invalid peer ID from bootnode, expected `QmQZ8TjTqeDj3ciwr93EJ95hxfDsb9pEYDizUAbWpigtQN` at address `/ip4/172.28.1.1/tcp/30333`.
dave            | 2020-04-21 13:44:43 💔 Invalid peer ID from bootnode, expected `QmQZ8TjTqeDj3ciwr93EJ95hxfDsb9pEYDizUAbWpigtQN` at address `/ip4/172.28.1.1/tcp/30333`.
charlie         | 2020-04-21 13:44:43 💔 Invalid peer ID from bootnode, expected `QmQZ8TjTqeDj3ciwr93EJ95hxfDsb9pEYDizUAbWpigtQN` at address `/ip4/172.28.1.1/tcp/30333`.

In that case, QmQZ8TjTqeDj3ciwr93EJ95hxfDsb9pEYDizUAbWpigtQN is not expected but the wrong one.

[bkchr]

No, the chain spec/cli provides the peer ids. This is the peer id that will be expected by the node and not the peer id that the other node returned.

So, the error message is correct and I will close this pr. Ty anyway.

[chevdor]

After discussion, an option would be to fix the message with a better one that address properly both cases.

Say we have 2 nodes: node1 and node2. For the sake of brevity, I keep the ids short below.

Node1 has a peer-id which n1.

When a user connects node2 to node1 and mentions that the peer-id should be node1 (which is wrong, it should be n1) , there are 2 possible explanations:

• the user providing the peer-id node1 made a mistake (that was my case)
• something malicious is going on and node1 is not the one we expect (the case initially implemented).

I will provide a new fix that provides a message that works for both cases.

[tomaka]

I took the liberty to merge master in this branch.