I cannot share the code, but in short something based on a payable fallback in a complicated multi contract application has this behavior:

1) tested in remix using VM: perfect

2) tested in remix on Ganache, win version both 1.22 and 2.00: crash on fallback activation (I.e. set 1 ether as value, then click on fallback label)

3) tested in remix on Ropsten: perfect

4) tested in remix on Rinkeby: perfect

No problems related to gas limit, as resulting to me at the moment.

Ganache has been the main test bed for the application until now. Now something is not right.

What could be the difference between those test bed that I can address here?

  • 2
    It happened to me a couple of times that a version of ganache-core was buggy. I'd try to write a minimum reproducible example (modifying metacoin box for example) and report it to github. As a workaround I'd try to test with a different version. Try to run your tests in a linux box, to discard a failed windows build. Without an example to test is impossibel to be more precise.
    – Ismael
    Dec 29 '18 at 3:45
  • @Ismael thank you, I appreciate your point. I’ll do, but I’m trying to ask for possible true differences to take into account. If the problem is simply a bug, your suggestion is the best.
    – Rick Park
    Dec 29 '18 at 7:45
  • Is there any codesize limit for a contract in ganache?
    – Rick Park
    Dec 29 '18 at 8:26
  • 1
    I add another case of “ok on remix-vm and ropsten, failing on ganache”: github.com/trufflesuite/ganache-core/issues/242
    – Rick Park
    Dec 29 '18 at 10:35

It seems that the difference be the code used for gas estimation, that is buggy. We wait for them to fix it!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.