frank

Unfortunately, the error still can be reproduced, but the frequency seems lower than before(6/100). @richard-ramos ``` messenger_raw_message_resend_test.go:292: response error: contact request not received messenger_raw_message_resend_test.go:298: Error Trace: /Users/frank/development/go/src/github.com/status-im/status-go/api/messenger_raw_message_resend_test.go:298 /Users/frank/development/go/src/github.com/status-im/status-go/api/messenger_raw_message_resend_test.go:260 /Users/frank/development/go/src/github.com/status-im/status-go/api/messenger_raw_message_resend_test.go:61 /Users/frank/development/go/src/github.com/status-im/status-go/vendor/github.com/stretchr/testify/suite/suite.go:187...

> I have run this a couple of times succesfully: > > ``` > go test -run TestMessengerRawMessageResendTestSuite/TestMessageSent github.com/status-im/status-go/api -count 100 -v -timeout 10000s > ``` > > please share...

more specific and smaller range logs: [cut_console.log](https://github.com/status-im/status-go/files/14953893/cut_console-output.log) [cut_geth.log](https://github.com/status-im/status-go/files/14953894/cut_geth.log) @richard-ramos

FYI, append these lines into `createAliceBobBackendAndLogin` fixed [issue](https://github.com/status-im/status-go/pull/4969#issuecomment-2044249115): ``` aliceWaku := s.aliceBackend.StatusNode().WakuV2Service() bobWaku := s.bobBackend.StatusNode().WakuV2Service() // NOTE: default MaxInterval is 10s, which is too short for the test err =...

WDYT? @status-im/status-go-guild cc @ilmotta

If `recover` is not used, how can the crash problems encountered by real users be located and solved? In the absence of logs, this will become very difficult. Real users...

Can `Sentry` ensure that the production environment won't have crashes and quickly locate the root cause of crashes for the PR build? If so, I'd love to use it! >...

> if the issue is logging, what about we catch the panic, log it, and throw it again? yeah, equal to `We can default to panic again!` 👍

because we can't trust other devs to fix panic when it panic, so basically, the code would be like following: ``` func safeGo(f func()) { go func() { defer func()...

> > if the issue is logging, what about we catch the panic, log it, and throw it again? > > But doesn't it log the panic by default? no,...