Adrian Cole
Adrian Cole
To fix this at servlet abstraction would be pretty gnarly because what we'd have to do is wrap the HttpServletRequest, so that we can later wrap `request.getAsyncContext().addListener(` This would also...
are there any unicode characters in your data? if so, where are they? ex. .name, .localEndpoint.serviceName, etc?
I know your data is sensitive. it might be easier for you to find the bug.. can you add code to ZipkinJsonV2JsonWriterTest like this, and test the fields you have...
sorry.. I believe there is a bug, but I can't see it anywhere. will need to know what is causing this and I can't currently find it cbf0bfea8
@jeqo if I recall correctly, there's a fix to this upstream now?
we should close this now with ActiveMQ release fixing this internally correct? meanwhile it is nearing a year old :D
As mongodb reactive is a different library, with a different version, we shouldn't embed it into brave-instrumentation-mongodb, rather brave-instrumentation-mongodb-reactivestreams this is similar to apachehttpclient and apachehttpasyncclient which have different versions...
Note: the reactive driver v4 is out. the current code is v3 and I'm not sure about its future. If someone knows what's the deal, let's try to brainstorm prior...
here's somewhat contrived code that breaks in exactly this fashion https://github.com/openzipkin/brave-webmvc-example/compare/trace-busters?expand=1
on the server-side I think the most sensible first step is create a child span when something else set a current span in scope. Even if there's double-instrumentation, at least...