You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Looking at its gas use, 21272, that's the base transaction fee (21000) + 4 byte of input data (4 * 64 gas), which is again consistent with a NOOP. I.e. this transaction executed correctly, and consumed a correct amount of gas, and reported correctly the success status.
That leaves the question, why does etherscan (any my local node too) report it as REVERTed. We're still looking into it, but our hunch is that the tracer code uses a slightly modified transaction execution pathway and somewhere the code of the contract was not flagged as deleted and still used that during tracing.
Yeah, so the error originates from the tracer which used some funky legacy code to dispose of deleted objects, not the consensus code. Swapping it out to the proper consensus function call makes everything run smoothly.
Sorry for the messup, I'll have to ping @mtbitcoin to re-trace after this is merged/released.
I have found interesting transaction on Rinkeby chain:
https://rinkeby.etherscan.io/tx/0x9d0f5ae03a1708f85858764d631a89ef373f92f759f4a6b773675ad02a8e6700
The transaction receipt indicates successful execution, while geth trace contains REVERT opcode. How it could be?
System information
Geth version: 1.8.4
OS & Version: Linux
The text was updated successfully, but these errors were encountered: