search

Node Operations

hashtag
Manage core services

This section provides a low-level explanation of how the services work, which is essential for debugging, even if the infrastructure is built using Kubernetes (K8s). Kubernetes also utilizes these methods under the hood.

circle-info

For deployment and management using Kubernetes (K8s), refer to this linkarrow-up-right

To ensure proper operation, the nodes must be started in the order specified in this document and shut down in the reverse order:

hashtag
op-geth

  • Start up the node

    • Archive Mode: It is crucial to run the op-geth node in archive mode using the -gcmode flag. Archive mode enables recording all historical data at any block number, which is particularly useful for recovery in case of issues.

    • WebSocket Support: Use the —ws flag to enable WebSocket. This allows DApps to listen for events on the L2 network.

    geth \
	--datadir=/db \
	--verbosity=3 \
	--http \
	--http.corsdomain=* \
	--http.vhosts=* \
	--http.addr=0.0.0.0 \
	--http.port=8545 \
	--http.api=web3,debug,eth,txpool,net,engine \
	--ws \
	--ws.addr=0.0.0.0 \
	--ws.port=8546 \
	--ws.origins=* \
	--ws.api=debug,eth,txpool,net,engine \
	--syncmode=full \
	--nodiscover \
	--maxpeers=0 \
	--networkid=901 \
	--rpc.allow-unprotected-txs \
	--authrpc.addr=0.0.0.0 \
	--authrpc.port=8551 \\
	--authrpc.vhosts=* \\
	--authrpc.jwtsecret=jwt-secret.txt \
	--gcmode=archive \
	--state.scheme=hash \
	--metrics \
	--metrics.addr=0.0.0.0 \
	--metrics.port=6060

  • Shut down the node

    • Using archive mode can help reduce data loss in case shutting down unexpectedly. However, the op-geth needs to be stopped gracefully by sending a SIGINT signal to the op-geth process.

    • The archive mode cannot address all scenarios, so it is beneficial to have another replica of op-geth and op-node running in sync with each other. This redundancy ensures that if one node goes down unexpectedly, the other can continue providing data and minimize downtime.

    • Some command examples for shutting down the op-geth node (all commands must to send SIGINT signals).

    # Kill process with a SIGINT directly
kill -2 <Process ID>

# 
docker stop -t 300 <Container ID>

hashtag
op-node

  • Start up the node

    • op-node processes L1’s data, communicates with op-geth and instructs op-geth on how to build L2.

    • Example command and result:

  • Shut down the node

    The op-node process is stateless, allowing it to be terminated using any method without risk of data loss.

hashtag
op-proposer

  • Start up the node

    • To activate the Fault Proof System, provide the contract address of the Fault Dispute Game Factory.

    • Ensure the --allow-non-finalized flag remains set to false on the mainnet (default value is false).

    • Enable admin option.

  • Shut down the node

    • The node process can be terminated using any method, as it does not require a specific shutdown procedure.

hashtag
op-batcher

  • Start up the node

    • Example command and result:

  • Shut down the node

    • We need to send this command first to the op-batcher process to stop the process gracefully.

    • Wait until the Batcher service stops then terminate the process:

hashtag
op-challenger

  • Start up the node

  • Shut down the node

    The op-challenger service is stateful, so to safely stop the process, you must send a SIGINT signal to the op-challenger process.

hashtag
How to check and update services

  • Check service configuration: Thanos stack’s services run by executing the binary file with configurable parameters or environment variables. So to understand the configuration, we need to check.

    • The parameters

    • The environment variables

PreviousArchitectureNextChain Configurations

Last updated