[OP Stack Deployment] Sequencer deployment: eth_sendRawTransaction method does not exist when submitting transactions into op-geth

[d-Bharti001]

Did you check the documentation?

• Yes

Did you check for duplicate questions?

• Yes

Issue Description

Sending transactions to the L2 op-geth RPC endpoint doesn't work. It says "the method eth_sendRawTransaction does not exist/is not available".

I deployed OP Stack components on top of a local blockchain spun up using Kurtosis.

So the structure is like this:

• L1 and L2 chain IDs are custom
• L1 has just one geth instance paired with a lighthouse instance
• L2 is deployed as per the deployment guide: https://docs.optimism.io/operators/chain-operators/deploy
• op-geth and op-node versions are compatible with each other

These are the versions used:

• op-deployer: v0.3.3
• L1 contracts: tag://op-contracts/v3.0.0
• L2 contracts: tag://op-contracts/v3.0.0
• op-geth: v1.101511.0 (when running --version of the executable: op-geth version 1.101511.0-rc.1-68075997-20250515)
• op-node: v1.13.3 (when running --version of the executable: op-node version v1.13.3-rc.1-b1e7c63b-1747318459)
• op-proposer: v1.10.0
• op-batcher: v1.13.1

I ran op-geth and op-node using the standard configurations given in the documentation: https://docs.optimism.io/operators/chain-operators/deploy/sequencer-node

Here are the commands for starting the components:

./bin/op-geth \
  --datadir=./op-geth-data \
  --http \
  --http.addr=0.0.0.0 \
  --http.port=$OP_GETH_HTTP_PORT \
  --http.vhosts="*" \
  --http.corsdomain="*" \
  --http.api=eth,net,web3,miner,debug,txpool,admin \
  --ws \
  --ws.addr=0.0.0.0 \
  --ws.port=$OP_GETH_WS_PORT \
  --ws.origins="*" \
  --ws.api=eth,net,web3,miner,debug,txpool,admin \
  --authrpc.addr=0.0.0.0 \
  --authrpc.port=$OP_GETH_AUTH_PORT \
  --authrpc.vhosts="*" \
  --authrpc.jwtsecret=$JWT_SECRET \
  --rollup.disabletxpoolgossip=true \
  --rollup.sequencerhttp=http://localhost:$OP_NODE_RPC_PORT \
  --syncmode=full \
  --gcmode=archive

./bin/op-node \
  --l1=$L1_RPC_URL \
  --l1.beacon=$L1_BEACON_URL \
  --l2=http://localhost:$OP_GETH_AUTH_PORT \
  --l2.jwt-secret=$JWT_SECRET \
  --rollup.config=.deployer/rollup.json \
  --sequencer.enabled=$SEQUENCER_ENABLED \
  --sequencer.stopped=$SEQUENCER_STOPPED \
  --sequencer.max-safe-lag=3600 \
  --verifier.l1-confs=4 \
  --p2p.listen.ip=0.0.0.0 \
  --p2p.listen.tcp=$P2P_LISTEN_PORT \
  --p2p.listen.udp=$P2P_LISTEN_PORT \
  --p2p.advertise.ip=$P2P_ADVERTISE_IP \
  --p2p.advertise.tcp=$P2P_LISTEN_PORT \
  --p2p.advertise.udp=$P2P_LISTEN_PORT \
  --p2p.sequencer.key=$PRIVATE_KEY \
  --rpc.addr=0.0.0.0 \
  --rpc.port=$OP_NODE_RPC_PORT \
  --rpc.enable-admin \
  --log.level=info \
  --log.format=json

(I checked: sequencer.enabled is actually true for op-node. I even tried adding some more flags to both op-geth and op-node, but there wasn't any resolution to the problem. All the env variables are pointing to the correct resources so far.)

• For sending transactions and checking balances, I used Foundry's cast (version 0.3.0)
• First I deposited some funds from L1 to L2 via. the l1StandardBridgeProxy contract. It worked.
• Then in L2, I tried to deploy a contract using Remix IDE. It showed "RPC error". I tried to send funds to another account (in L2, using cast send)... it threw this error: "the method eth_sendRawTransaction does not exist/is not available"

I did some debugging:

• I checked the code of op-geth, and found that it's sending eth_sendRawTransaction calls to seqRPCService (which points to the rpc endpoint of op-node, which is set using the flag: --rollup.sequencerhttp = http://localhost:$OP_NODE_RPC_PORT).
• In the function op-geth :: eth/api_backend.go :: func (b *EthAPIBackend) SendTx(ctx context.Context, signedTx *types.Transaction) error, it has a line b.eth.seqRPCService.CallContext(ctx, nil, "eth_sendRawTransaction", hexutil.Encode(data)).
  • That function calls c.sendHTTP(ctx, op, msg) which works but there might be a problem later in the line where it calls op.wait(ctx, c).

I need to know what I have to change in this setup to make it work. I've been trying since two days. I really need help here 🙏🏻

Logs

No response

Additional Information

No response

Feedback

No response

[d-Bharti001]

Solved it!

op-geth should not redirect the transaction submissions to op-node.

Removed --rollup.sequencerhttp while launching op-geth, and it worked.

[hash1go]

cc: @opfocus
Could you please close this ticket?