Closed YuXiaoCoder closed 1 month ago
I don't see "error fetching MaxDataSize from sequencer inbox" anywhere in the prysm codebase. Additionally, the rpc call you have provided doesn't make sense in the context of eth mainnet: it is a call from the zero address to an arbitrum contract.
Are you trying to run an Arbitrum node?
Edit: yes, the error looks to originate from arbitrum nitro which is meant to be coupled with Geth.
I assume "the node responds normally" means that the eth_call
succeeds which doesn't make sense to me. why would it succeed when the gas cap is set to the 50_000_000 default, and not when it is infinite... Maybe others will have more insight on this.
Yes, I'm running arbitrum nitro.
I found the bug, in the doCall method in internal/ethapi/api.go, it was previously setting AddGas as
gp := new(core.GasPool).AddGas(math.MaxUint64)
the new code changed to new(core.GasPool).AddGas(globalGasCap)
, if we set RPCGasCap(globalGasCap) to 0, we intended to set infinite If we set RPCGasCap(globalGasCap) to 0, it is meant to be infinite.
This bug can be bypassed by changing the run command
/opt/ethmain/core/geth --config=/mnt/ethmain/conf/config.toml --rpc.gascap=9223372036854775807 --rpc.txfeecap=0
Experiencing similar issues.
Assuming that is related to https://github.com/ethereum/go-ethereum/pull/27720
System information
Geth version:
1.14.9
CL client & version:Prysm@v5.1.0
OS & Version:Linux
Expected behaviour
eth_call returns the response normally
Actual behaviour
Steps to reproduce the behaviour
Prysm Command:
Geth Command:
The version of Geth before we upgraded was 1.14.5,When I remove the
--rpc.gascap=0 --rpc.txfeecap=0
parameter, the node responds normally, but the 0 value is unlimited in meaning ah!