evm-bully fails on Rinkeby testnet, block 55, tx 0

[frankbraun]

Transaction Id 361165UkrpXsBPxFuvqG1UA97camS1CZYZc6mb7KjPSE
evm-bully: error: failure:
{
  "ActionError": {
    "index": 0,
    "kind": {
      "FunctionCallError": {
        "ExecutionError": "Smart contract panicked: ERR_OUT_OF_FUND"
      }
    }
  }
}

See replay failing transactions on how to replay rinkeby-block-55-tx-0.tar.gz

[frankbraun]

See also #2.

[birchmd]

This one appears to be an issue with the rinkeby genesis state initialization with the bully. The failing tx is the very first one on the chain and it is a simple transfer from one address to another. The source address in that tx is 0x31b98d14007bdee637298086988a0bbd31184523, which is funded with effectively infinite ether at genesis (this can be seen on etherscan).

In the Aurora state provided to repro the failure, that address has a balance of only 2 Wei, which is incorrect based on the genesis state. This is why I think the problem was with the initialization. Assigning this one back to @frankbraun to investigate.

[frankbraun]

@birchmd You were right in the discussion in #30, RawU256 must be big-endian (what the EVM uses) and not little-endian (what WebAssembly uses). So the balance was encoded wrongle by the evm-bully (fixed in aurora-is-near/evm-bully@7705c27) and that caused this error.

[frankbraun]

Still fails on newest HEADs of nearcore and aurora-engine (master): rinkeby-block-55-tx-0.tar.gz

But with different error message:

{
  "Failure": {
    "ActionError": {
      "index": 0,
      "kind": {
        "FunctionCallError": {
          "ExecutionError": "Smart contract panicked: Failed read storage [src/connector.rs:69:71]"
        }
      }
    }
  }
}

Ropsten and Görli now also fail with the same error message on the first transaction.

Might be related to #2.

[birchmd]

This is now resolved. Rinkeby now makes it all the way to block 34300 (see https://github.com/aurora-is-near/aurora-engine/runs/3971726940?check_suite_focus=true )