-
Notifications
You must be signed in to change notification settings - Fork 71
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Consensus tests fail in AppVeyor intermittently #305
Comments
Seems like the test name is constant. https://ci.appveyor.com/project/torquem/silkworm/builds/40115019
|
Reproduced locally. Launched
|
Added
|
Another
Chain changes but the test file is always the same |
Running a batch of 256 consecutive tests on branch |
Hmmm not looking good. Less than 50 tests and already >10% failures
|
So the batch completed. 256 attempts and 25 failures
|
I just modded
Hence it might not be a memory problem or something related to directory recursion order |
Amendment of the above ... i run a full batch of 4096
The failure ratio is way lower than executing a fully fledged Edit ... just ran the very same test on Linux (well ... WSL) and not a single failure on 10k runs |
Definitely an MSVC problem which is somewhere into Transactions are all executed (number of receipts matches number of transactions) but for some reason gas_used is computed (intermittently) wrong |
Interestingly the failures on Berlin network give different
|
My gut feelings lean towards some issue related to MSVC with intx |
Under |
Have you tried running it under Address and Undefined Behaviour Sanitizers? |
Not yet. Returning in office now and will do |
Crap. If I try to build MSVC with address sanitizer ABSEIL complains it does not find his own address sanitizer
|
For example,
randomStatetest445_d0g0v0_Berlin
in https://ci.appveyor.com/project/torquem/silkworm/builds/40042149The text was updated successfully, but these errors were encountered: