Skip to content
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

Exception semantic conventions need to be marked stable #3192

Closed
trask opened this issue Feb 9, 2023 · 5 comments
Closed

Exception semantic conventions need to be marked stable #3192

trask opened this issue Feb 9, 2023 · 5 comments
Assignees
Labels
area:semantic-conventions Related to semantic conventions spec:trace Related to the specification/trace directory

Comments

@trask
Copy link
Member

trask commented Feb 9, 2023

Opening for now primarily to discuss/track whether this is a blocker for marking HTTP semantic conventions stable.

My initial thought is that we won't be able to declare HTTP instrumentations stable without this, and therefore this is a blocker for HTTP semantic convention stability.

@trask
Copy link
Member Author

trask commented Feb 9, 2023

from @jsuereth

My main fear with current exception encoding is inetraction with the length-limit on the actual exception stacktrace string, and whether that will lead to poor experience in debugging exceptions from OTEL

@jsuereth
Copy link
Contributor

jsuereth commented Feb 9, 2023

Adding some more thoughts looking at specifics:

  • We should add a spec compliance metrics where each language SiG can denote whether they have documented their exception.stacktrace format. This is implicitly required in the current convention. See this recommendation as an option for language SIGs.
  • I'm worried about the interaction of OTEL_ATTRIBUTE_VALUE_LENGTH_LIMIT and exception.stacktrace. I'm not sure users of the value length limit would think about stacktraces as an attribute. We should entertain if we want to have some special interaction of those two or just encourage users not to set length limits, as the current default is unlimited.
  • Should we ask language SiGs to update the semconv specification document with their official layout? This way the documentation for how to leverage the field is centrally located.

@reyang
Copy link
Member

reyang commented Feb 11, 2023

My vote: it's not blocking HTTP semantic convention.
Reason:

  1. exception is not going to be part of the core HTTP metrics (so there is no concern of breaking the metrics streams)
  2. exception stack should be opt-in due to the size and potential sensitive information

@trask
Copy link
Member Author

trask commented Feb 13, 2023

until exception semantic conventions are marked stable, we cannot mark any Java HTTP instrumentations as stable.

but I agree this should not be block HTTP semantic conventions themselves from being marked stable.

@lmolkova
Copy link
Contributor

lmolkova commented Jul 9, 2024

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area:semantic-conventions Related to semantic conventions spec:trace Related to the specification/trace directory
Development

No branches or pull requests

5 participants