Interesting Issue About OpenInvariantsTest

[codememeWY]

assert(wethValue + wbtcValue > totalsupply);
if i use > it will trigger Ran 1 test for test/fuzz/OpenInvariantsTest.t.sol:OpenInvariantsTest
[FAIL: failed to set up invariant testing environment: panic: assertion failed (0x01)] invariant_protocolMustHaveMoreValueThanTotalSupply() (runs: 0, calls: 0, reverts: 0)
but if i change it to >=
boom it passed the test
Can somebody told me whats the reason

[EngrPips]

What seems logical is the assertion [assert(wethValue + wbtcValue > totalsupply);] is being checked when no weth and wbtc collateral have been deposited, and also no stable coin has been minted which means totalsupply is also 0 so the condition assert(wethValue + wbtcValue > totalsupply); which will then be assert(0 + 0 > 0) is false so the test breaks and the set was not able to finish it's setup but when the assertion assert(wethValue + wbtcValue >= totalsupply);then it becomesassert(0 + 0 >= 0)` which is true then the test process is able to continue.

But the question here is, does the Fuzz test run all of its processes for the first run and never make a successful deposit to the protocol, and also never mint any dsc? If yes, then why is it saying failed to set up invariant testing environment because obviously for the Handler to run, then the invariant testing environment must have been successfully set up, or is it giving failed to set up invariant testing environment because it is the first assertion that is done after running a fuzzing process, If yes then that is a confusing error as it is giving the impression that fuzz testing setup didn't complete which isn't true. In that case then we can say Foundry should work on a better error message.

Another scenario could be that it is first asserts assert(wethValue + wbtcValue > totalsupply); before running any fuzzing process. If that is the case, then I don't think that is the correct order of a fuzzing process, and we should report to Foundry.

You can do more personal research and find out what is actually going on, and maybe report to Foundry if you find out that it is an issue from Foundry but if you find out that it is an issue with the way we set up our fuzzing test suite then we will be glad to hear how could have better set it up from you. This is an interesting issue to dig into, but I won't because I have plenty of other priorities but if you think you have the time to dig and find out, best of luck, Friend.

[MaYatKit]

I have same issue.
It supposes to print logs weth value: 0, wbtc value: 0, etc like patrick's codes.
But I got [FAIL: failed to set up invariant testing environment: panic: assertion failed (0x01)]...
I know it may caused by Foundry bug or whatnot, I can just ignore it and keep go ahead, but it does trigger my OCD :)

[EngrPips]

Yeah, sometimes we just want to figure out why things break.