ci(-race): race condition with -race flag

[DIGIX666]

-race flag

I do going to try make this issues #2097

This is my draft PR for -race flag. For moment not all files is fixet.
My plan for solve these race conditions :

1. Run go test -race ./...

2. Check all files in running DATA RACE and Goroutine
   
   

3. Use a best technique (mutex, chan, sync/atomic) for solve Data Race and Goroutine

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 60.85%. Comparing base (01ee5a9) to head (f0eb99a).
Report is 18 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #2663   +/-   ##
=======================================
  Coverage   60.84%   60.85%           
=======================================
  Files         563      563           
  Lines       75190    75184    -6     
=======================================
  Hits        45753    45753           
+ Misses      26068    26063    -5     
+ Partials     3369     3368    -1     

Flag	Coverage Δ
contribs/gnofaucet	15.31% <ø> (+0.85%)	⬆️
gnovm	65.63% <ø> (ø)
misc/genstd	80.54% <ø> (ø)
misc/logos	20.23% <ø> (+0.35%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[DIGIX666]

I am currently correcting the test errors on the bytes_test.gno, L542 and sort_test.gno, L656 files but I don't understand if // XXX remove testenv means that it is no longer used or if it should be replaced because testenv doesn't exist in gno and I wanted to replace it with os.Getenv but it doesn't exist on gno either because I wanted to use an export RACE=true in the CI test-with-race

cc @ajnavarro @thehowl

[DIGIX666]

Flag Update

I've removed the t.Parallel() in files with concurrency but I think we should remove all t.Parallel() (I'll try before I push) to reduce the risk of race conditions but also added Mutex.

gnovm

The issue with the test go test -race -timeout 30m ./... in the gnovm folder is that there's an error every time, even without a timeout, on the file xfactor_long.gno because it panics.

panic: test timed out after 10m0s
running tests:
	TestFilesNative (10m0s)
	TestFilesNative/xfactor_long.gno (1m11s)

I tried to skip the file during the test by adding:

if filepath.Base(path) == "xfactor_long.gno" {
	t.Skip("skipping non-gno file:", path)
}

in the file eval_test.go L14

But it didn’t change anything.

For the gnovm folder, I think the -race flag is not relevant because the tests in /gnovm/tests/files are .gno files that intentionally test for created errors.

tm2

Currently for tm2, the remaining test is TestMempoolNoProgressUntilTxsAvailable, which times out, and the test TestNodeDelayedStart, which gives me the following error, but I don’t know how to fix it:

Data: failed to listen on 0.0.0.0:26657: listen tcp 0.0.0.0:26657: bind: address already in use

For the CI gno.land/Run Main/Go Test/test and tm2/Run Main/Go Test/test, I don’t understand their failures because I tested them locally and they passed.

Thank you for checking my changes when you have the time and for guiding me if possible to fix the remaining issues :)

[mvertes]

Thank you for your contribution, but this is part where we are working on refactoring the whole set of tests, also in relation to gnovm, it's better to let this to the core team. There is a lot missing context for you, and it's up to us to clarify before welcoming contributions on this specific part, thanks.

[DIGIX666]

Thank you for your contribution, but this is part where we are working on refactoring the whole set of tests, also in relation to gnovm, it's better to let this to the core team. There is a lot missing context for you, and it's up to us to clarify before welcoming contributions on this specific part, thanks.

Thanks for your feedback and explanation I understand what you mean. In any case, it would have allowed me to learn something new 🙏🏼