Milkomeda EVM Passive

Milkomeda EVM node configurations for partners wanting to connect to the evm nodes p2p network

NOTE: Besu supports two node types, commonly referred to as Full nodes and Archive nodes. This setup runs an Archive Node by default.

• Usage
  • Start Script
  • Docker Compose Direct
• Config
  • Change to Full Node
• Troubleshooting
  • Not Syncing
  • Permission Denied
  • Known Errors for this remediation

Usage

Start Script

There is a simple start script provided to make starting/stoping/cleaning up easier. By default it will launch the EVM in c1-devnet.

NOTE: Add the -o flag to all commands if using WSL & MacOS (disables Host Network mode and Set Ports)

• Script Help for the most up to date commands ./start.sh -h
• Find Network(s) to start: ./start.sh -l
• Start EVM: ./start.sh -n ${NETWORK} -r
  • Add Log Tailing: ./start.sh -n ${NETWORK} -r -t
• Stop EVM: ./start.sh -n ${NETWORK} -s
• Tail logs: ./start.sh -n ${NETWORK} -t
• Cleanup EVM: ./start.sh -n ${NETWORK} -c

Docker Compose Direct

NOTE: Use docker-compose-non-linux.yml instead for WSL & MacOS usage (disables Host Network mode and Set Ports)

• Switch to Directory of network you wish to Run EVM in Example: cd c1-devnet
• Start EVM:
  • docker-compose -f docker-compose.yml -p ${NETWORK}-passive build
  • docker-compose -f docker-compose.yml -p ${NETWORK}-passive up --detach --force-recreate
• Stop EVM: docker-compose -f docker-compose.yml -p ${NETWORK}-passive stop
• Tail logs: docker-compose -f docker-compose.yml -p ${NETWORK}-passive logs --follow --tail="all"
• Cleanup EVM: docker-compose -f docker-compose.yml -p ${NETWORK}-passive down

Config

Change to Full Node

NOTE: If you enable sync-mode="FAST" then you need 5 peers before it'll begin syncing. The setting to modify this is below

• Edit the config file at ${NETWORK}/config/config.toml
  • Change sync-mode="FULL" to sync-mode="FAST"
  • Optional: Add number of minimum peers to config file (e.g.; fast-sync-min-peers=2). Also can be added to docker-compose.yml under the Command section as --fast-sync-min-peers=2

Using env file:

Following settings can be customized using .env file. Example:

### Besu repository (mainnet default: hyperledger/besu; devnet default: dcspark/besu)
BESU_REPO=dcspark/besu
### Besu tag (mainnet default: 22.1.3; devnet default: 22.10.1-milkomeda-c1)
BESU_TAG=22.1.3
### Besu data dir (default: ./storage/data)
DATA_DIR=./storage/data
### Besu data dir (default: ./storage/logs)
LOGS_DIR=./storage/logs

.env file has to be pleaced in ./c1-mainnet or ./c1-devnet directory

Troubleshooting

Not Syncing

1. Pull Latest Updates from Github
2. Trying restarting the service
   • docker-compose -f docker-compose.yml -p ${NETWORK}-passive restart
3. If issue persists, run the commands under the Usage section in this order Stop EVM Clean EVM Start EVM

• Disconnect - Inbound - 0x04 TOO_MANY_PEERS

  If you have on logs the following message over and over, it means that the bootstrap nodes are full and there are no other nodes to connect too at the moment.

{"timestamp":"2022-03-29T13:54:32,453","container":"xxxx","level":"INFO","thread":"EthScheduler-Timer-0","class":"FullSyncTargetManager","message":"No sync target, waiting for peers: 0","throwable":""}

To avoid peer starvation and help others to connect to p2p network, the users should enable p2p discovery and have their nodes accessible so the p2p network doesn't rely only on bootstrap/static nodes.

• https://besu.hyperledger.org/stable/private-networks/how-to/configure/bootnodes
• https://besu.hyperledger.org/stable/public-networks/how-to/connect/configure-ports
• https://besu.hyperledger.org/stable/public-networks/how-to/connect/specify-nat

Full Node Stops Syncing with "no peers"

This is a known problem with the Besu full node where for some reason it loses known static peers and it's not able to recover. During full sync this may happen a few times and it's resolved by restarting the node e.g. docker restart XXX

Permission Denied

If you get a Permission Denied error, e.g.:

root@blockscout:~/milkomeda-evm-passive# docker-compose -f c1-devnet/docker-compose.yml -p c1-devnet-passive up
[+] Running 1/0
 ⠿ Container c1-devnet-passive-besu-1  Created                                                                                                                                                                              0.0s
Attaching to c1-devnet-passive-besu-1
c1-devnet-passive-besu-1  | 2022-04-05 14:14:16,508 main ERROR Unable to create file /tmp/besu/besu-blockscout.log java.io.IOException: Permission denied

run the following command(s),

• chmod a+w -R ${NETWORK}/storage
  • You may have to restart docker-compose -f docker-compose.yml -p ${NETWORK}-passive restart

Known Errors for this remediation

ANTLR Runtime version 4.7.1 used for parser compilation does not match the current runtime version 4.8/opt/besu/data/DATABASE_METADATA.json (Permission denied)