[Bug] MegaETH - Testnet local node returns 400 on eth_getBlockByHash

[Luscha]

Bug report

I'm trying to run a local node v0.38.0 using the official docker compose for testing purpose indexing MegaETH Testnet.
I'm using the official RPC endpoint in the docker compose and deplying a subgraph.

The said subgraph works in the managed environment but gives 400 when trying to call eth_getBlockByHash in my local deployment.

I tried to curl the RPC directly and, in fact, it retuns a 400 because MegaETH do not support full block request.
I haven't found anything in the official documentation about how to properly configure a localnode against MegaETH.

The error appears before deploying the subgraph, when the node starts-up.

---

Docker compose:

version: '3'
services:
  graph-node:
    image: graphprotocol/graph-node
    restart: "no"
    ports:
      - '8000:8000'
      - '8001:8001'
      - '8020:8020'
      - '8030:8030'
      - '8040:8040'
    depends_on:
      - ipfs
      - postgres
    extra_hosts:
      - host.docker.internal:host-gateway
    environment:
      postgres_host: postgres
      postgres_user: graph-node
      postgres_pass: let-me-in
      postgres_db: graph-node
      ipfs: 'ipfs:5001'
      ethereum: 'megaeth-testnet::https://carrot.megaeth.com/rpc' #removed archive capability
      GRAPH_LOG: debug

  ipfs:
    image: ipfs/kubo:v0.17.0
    restart: "no"
    ports:
      - '5001:5001'
    volumes:
      - ./data/ipfs:/data/ipfs:Z

  postgres:
    image: postgres
    restart: "no"
    ports:
      - '5432:5432'
    command:
      [
        "postgres",
        "-cshared_preload_libraries=pg_stat_statements",
        "-cmax_connections=200"
      ]
    environment:
      POSTGRES_USER: graph-node
      POSTGRES_PASSWORD: let-me-in
      POSTGRES_DB: graph-node
      PGDATA: "/var/lib/postgresql/data"
      POSTGRES_INITDB_ARGS: "-E UTF8 --locale=C"
    volumes:
      - ./data/postgres:/var/lib/postgresql/data:Z

Subgraph manifest:

specVersion: 0.0.6
schema:
  file: ./schema.graphql

dataSources:
  - kind: ethereum/contract
    name: GameFactory
    network: megaeth-testnet
    source:
      abi: GameFactory
      address: "0x461733951532b44E229d52AFc8756729e2967959"
    mapping:
      kind: ethereum/events
      apiVersion: 0.0.8
      language: wasm/assemblyscript
      file: ./src/mappings/factory.ts
      entities:
        - Game
        - Factory
      abis:
        .....

Relevant log output

Local Node output:

Jul 13 21:55:02.422 DEBG Trying again after eth_getBlockByHash RPC call for block hash 0xf365c3a3030b04646d93040e009ac61d0724b8466d703d64fd174cc18b6ad514 failed (attempt #8) with result Err(code 400), provider: megaeth-testnet-rpc-0, component: EthereumPollingBlockIngestor

Curl output against https://carrot.megaeth.com/rpc

>> curl -X POST https://carrot.megaeth.com/rpc \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc":"2.0",
    "method":"eth_getBlockByHash",
    "params": ["0x92e19450198a1320d581c9b461a195a8939fd0619a5992a3f2970ead75d71b3d", true],
    "id": 1
  }'
{"jsonrpc":"2.0","error":{"code":-32602,"message":"full block not allowed"},"id":null}

IPFS hash

No response

Subgraph name or link to explorer

No response

Some information to help us out

• Tick this box if this bug is caused by a regression found in the latest release.
• Tick this box if this bug is specific to the hosted service.
• I have searched the issue tracker to make sure this issue is not a duplicate.

OS information

Other (please specify in your bug report)

[shiyasmohd]

Hey @Luscha
This is an RPC issue and that curl request should be working in order to graph node to run successfully. Archive capability for the rpc is required.

[Luscha]

@shiyasmohd That's what i've thought, the fact is that i was able to deploy and index MegaETH test net using thegraph.com meaning the managed service is either able to workaround the RPC limitation or uses another RPC endpoint.

I was wondering if this is a choice / limitation of the self-hosted vs managed version of theGraph or a missconfiguration.

Not being able to run the node locally has a huge impact in the developer experience since i have to wait hours every time for TheGrap to index the chain (it uses a slower worker pool than the local instance)

[shiyasmohd]

Infact this is an issue related to the RPC you use in graph node locally, If you can use an RPC that support archive calls, it will be running locally.