@kezhenxu94 I think with using graceful shutting down, we need a simple e2e to verify whether OAP and UI could be shut down successfully. Because, once we have one unexpected thread, the shutdown wouldn't work.

Originally posted by @wu-sheng in apache/skywalking#7020 (comment)

Requirement - what kind of business use case are you trying to solve?

Hello team, in PR jaegertracing/jaeger#2798 the tls connection is enabled in collector, it would be good to support configuring cipher for it.

It would be similar to the fluentd tls settings in https://docs.fluentd.org/configuration/transport-section#tls-setting, you can config the ciphers fi

Current Behaviour

The flamegraph sorts the siblings in ascending order by frame's name.
The reference to code is here

Expected Behaviour

The siblings should appear in order of their start time.
This can be achieved according to library docs mentioned [here](https://github.com/spiermar/d3-fla

#750 added Hedged requests to GCS/S3/Azure backends.

There was a discussion on the PR to add metrics for tracking the performance of hedged requests, and the corresponding stats have been introduced in the hedgedhttp library in the latest release - https://github.com/cristalhq/hedgedhttp/releases/tag/v0.6.0

Use these stats to generate prometheus metrics for each backend.

After the merge of this PR: open-telemetry/opentelemetry-dotnet#1548

We should create more unit tests just like open-telemetry/opentelemetry-dotnet#1537 to validate the full behavior.

Tracking issue, similar to jaegertracing/jaeger#2645

Done:

• #548
• #549
• #550

Remaining:

• publish crossdock images
• remove .travis.yml
• update build badge in README
• change required checks to new GH jobs

The current getting started guide assumes the user will be starting using the SDK building block packages like tracing, node, and metrics. In the future, it is expected most users will use the SDK package as a starting point and it would be better for the getting started guide to reflect that.

Cannot use same trace ID accross different node of microservices

On the document of https://opentracing.io/docs/overview/tracers/#propagating-a-trace-with-inject-extract, there is a concept of inject and extract. This is working fine with http carrier. Yet, I cannot reproduce this with Text Map carrier.

Please create a doc for u

@thegryfon noticed that in community axios implementations, some leak contexts and you wouldn't necessarily know until a second root trace is attempted.

here's an example scenario I think tests could cover:

there is no parent trace in scope
create a client request
create a second client request
verify that both client requests are root spans

Is your feature request related to a problem? Please describe.

The .. module has a conditional dependency on the parseurl library. This library is not an agent dependency. However, if parseurl is present then the agent will use it instead of it's own home grown URL parsing.

We've had users report that this conditional `re