-
Notifications
You must be signed in to change notification settings - Fork 46
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
re-uses w3c traceparent parser from brave #532
re-uses w3c traceparent parser from brave #532
Conversation
Signed-off-by: Adrian Cole <[email protected]>
...er-tracing-bridge-brave/src/main/java/io/micrometer/tracing/brave/bridge/W3CPropagation.java
Show resolved
Hide resolved
Signed-off-by: Adrian Cole <[email protected]>
Signed-off-by: Adrian Cole <[email protected]>
note this also uses |
AFAIR W3C + Brave is the default propagation format in Boot and people are using this with messaging successfully 🤷 , regardless I'm all for not having to support additional code and if all the tests are passing I'm all for merging this |
thanks for merging. I made a note again about the uber bug here, you can decide when to address it as it is 100pct an uber bug and in no way subjective #532 (comment) |
ps the reason this never happens in brave hosted instrumentation is because there is a kind-specific means to inject and extract stuff, as well how to create the consumer span. The upstream code will ignore the shared flag (e.g. never use joinSpan even if it is shared). The problem is that 3rd party instrumentation may not be careful like this. Particularly, sometimes things I don't remember spring integration or something can be confused about what kind the span is as the abstraction can wash away the details. The best way to prevent issues is to be context sensitive and put in tests that consumer side (span kind == consumer) never attempt to join a span. They might be here, but yeah setting shared unconditionally as this issue shows is a smell. EOF and last I'll talk about this promise! I hope it helps future somebody. |
This also highlights a very significant behavior bug which implies folks aren't using this with messaging, or using B3 instead.