🐛 Bug Report

The TestTimeout test is sometimes failing randomly.

To Reproduce

$ go test -run TestTimeout -count 100 ./examples/internal/integration

Expected behavior

Test not to be flaky

Actual Behavior

The tests fails with an incorrect status code error:

integration_test.go:1395: resp.StatusCode = 500; want 504

See https://app.circle

我需要

可我在hyperf中不知道该如何使用write

In our codebase, we push RequestContext and immediately run some code with try-with-resources.
For example:

try (SafeCloseable ignored = ctx.push()) {
    logger.trace(decorate(msg));
}

If RequestContext provides run(Runnable) or call(Callable) we can reduce boilerplate code and simplify it.

ctx.run(() -> logger.trace(decorate(msg));

This is inspired by g

virtual host options

We should also hook up a grpc-go client to test the interop with the tonic server.

Currently, in https://github.com/hyperium/tonic/blob/master/interop/test.sh we only test tonic client -> grpc-go server and tonic client to tonic server. We should add an additional client test that will go from a grpc-go client to tonic server.

Issue Description

是否有计划用actions 完全取代 TravisCI
感觉actions 速度上比 TravisCI更好点 ，另外也是github原生的。

Type: feature request

Describe what feature

In integration tests today we don't test properly that reset API wipes the node, we just test for reboot.

We could have used filesystem UUID change to verify that node got wiped and re-installed. (thanks @Ulexus)