https://github.com/openzipkin/zipkin/blob/master/zipkin/src/main/java/zipkin2/Span.java#L759

tags = builder.tags == null ? Collections.emptyMap() : new LinkedHashMap<>(builder.tags);

The empty map is immutable and makes it difficult to add a tag to a already created span. This will be helpful with performance. Otherwise, have to turn it in to builder and back to a Span when this type is us

There was a huge performance problem when using sleuth in the reactive stack. See here

This was partially addressed by this

But the downside 1478 of contents of MDC would be lost anytime event

Currently, responses with status code <200 are being marked as errors.

https://github.com/openzipkin/zipkin-js/blob/master/packages/zipkin/src/instrumentation/httpServer.js#L91
https://github.com/openzipkin/zipkin-js/blob/master/packages/zipkin/src/instrumentation/httpClient.js#L43

Comes from openzipkin/brave#874

Ping @adriancole @ghermeto

Often it is useful to add remoteEndpoint to the client spans. Specifically http requests.

Currently nor the client or the transport support this functionality. The most logical options that comes to my mind is that we add an option for both.

Ping @basvanbeek @adriancole

Hinting from the below comment and an encoding bug I think flags needs to be revisited:

        public const string Sampled = "X-B3-Sampled"; // Will be replaced by Flags in the future releases of Finagle

The X-B3-Flags header back maybe 2012 or so was thinking about becoming more, but never did. Reasons including awkward overlap with sampled, and also it is mainly covering binar

Test assertions in https://github.com/jcchavezs/zipkin-php/blob/master/tests/Unit/Reporters/HttpTest.php#L41 can be improved like in https://github.com/DataDog/dd-trace-php/pull/37/files#r202176688

Other tracer implementations include a setting to force client and server spans to have different spanId's. In Go the setting is WithSharedSpans(false) and in Java it's supportsJoin(false).

There should be something s

I am trying to deploy the docker image gcr.io/stackdriver-trace-docker/zipkin-collector inside kubernetes cluster. I was wondering what would be the best endpoint to have the liveProve / readyProbe functionality for a kubernetes pod.

The SQS instrumentation for brave uses queue.url where the XRay storage converter expects aws.queue_url

I think the XRay encoder should support braves tags in addition to the XRay named counterparts.

In the last release vote, it seemed evident the README could be clarified in a couple ways.

• beyond examples, further clarify that proto and swagger in this project are inputs to other tools
  • ex. point to the main docs of the IDL compiler
  • note that most users will not use these directly as tracing libraries will
• address a frequently asked question which is "why is there a maven bui

xray is here, but the amazon daemon is not present on the image. This means it is probably not a valid setup. AFAICT can either remove the xray module from this image, or add the x-ray daemon to it with instructions for how to configure it.

https://github.com/openzipkin/zipkin-aws/tree/master/autoconfigure/storage-xray#configuration