Node Connection Status

[m-janos]

Checklist

• I have followed (at least) the PR section of the contributing guide.
• I fixed all necessary PR warnings
• The commit history is clean
• The E2E tests are passing
• If possible, the issue has been divided into more subtasks
• I did a self review before requesting a review from another team member

Description

• Display connection status of each node in the network list (/nodes page)
• Used getpeerinfo to retrieve all connected peers, compare with node list to determine status

Test Plan

1. in scripts/development/docker-compose.yml add following service right after master-node:

 slave-node:
   build:
     context: ../../blockchain
   environment:
     ORGANIZATION: ${ORGANIZATION}
     RPC_PORT: ${RPC_PORT}
     RPC_PASSWORD: ${RPC_PASSWORD}
     P2P_HOST: master-node
     P2P_PORT: ${P2P_PORT}
     API_PROTO: http
     API_HOST: master-api
     API_PORT: ${API_PORT}
     EXTERNAL_IP: master-node
   volumes:
     - /tmp/bcSlaveVolume:/root/.multichain
   ports:
     - "7448:7447"

2. Run app as normal, approve new node and check that Network table on /nodes page displays two nodes with Connection status
3. In new terminal window run docker exec -it trubudget-dev_slave-node_1 bash
4. In afterwards-opened subshell run multichain-cli TrubudgetChain stop
5. Check frontend, one node should be disconnected
6. Return to the subshell and run multichaind -txindex -port=7447 -autosubscribe=streams TrubudgetChain@master-node:7447 “” -externalip=master-node
7. Refresh the page on the frontend. The node should be connected again

Resources

• Multichain-CLI tutorial

Issue

Subtask of #746

[openkfwCI]

	Notes	Time
✅	Note for Reviewer: E2E tests on remote server succeeded	Tue, 23 Feb 2021 14:19:55 +0000

Generated by E2E-Test

[georgimld]

Found a possible bug:
When a slave node is trying to connect to the network, it is retrying every 10 seconds and the logs in the container are

"Please ask blockchain admin or user having activate permission to let you connect and/or transact:
multichain-cli TrubudgetChain grant addressOfNode connect
multichain-cli TrubudgetChain grant addressOfNode connect,send,receive"

When approving a new node (Step 2 in the test plan), the node is given permissions (check using the multichain-cli command listpermissions * addressOfNode), but the node still appears offline (getpeerinfo shows no node) and according to the logs in the node's container the slave node keeps restarting every 10 seconds - which is why it appears not connected, even though it does have the necessary permissions to connect.

[m-janos]

Found a possible bug:
When a slave node is trying to connect to the network, it is retrying every 10 seconds and the logs in the container are

"Please ask blockchain admin or user having activate permission to let you connect and/or transact:
multichain-cli TrubudgetChain grant addressOfNode connect
multichain-cli TrubudgetChain grant addressOfNode connect,send,receive"

When approving a new node (Step 2 in the test plan), the node is given permissions (check using the multichain-cli command listpermissions * addressOfNode), but the node still appears offline (getpeerinfo shows no node) and according to the logs in the node's container the slave node keeps restarting every 10 seconds - which is why it appears not connected, even though it does have the necessary permissions to connect.

A new node is registered as connected after some time (one case takes 14 Minutes) and the "Please ask blockchain admin" message will disappear by itself.