-
Notifications
You must be signed in to change notification settings - Fork 1.5k
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
[service] implement a noop tracer provider #11026
[service] implement a noop tracer provider #11026
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is there any upstream issue about the
memory concerns around the SDK's noop tracer provider
?
I attended the go SIG on 28-Aug-2024, and the end result of the discussion was that because the no-op tracer still has to propagate context, the proposed solution if we need an alternative was to implement our own no-op tracer provider. |
This is to address some of the memory concerns around the SDK's noop tracer provider. Instead of using the published noop tracer provider, we implement our own to avoid the cost of context propagation, which has been proven to be high enough to cause issues for end users. Part of open-telemetry#10858 Signed-off-by: Alex Boten <[email protected]>
Signed-off-by: Alex Boten <[email protected]>
94b236d
to
c7197a8
Compare
Signed-off-by: Alex Boten <[email protected]>
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #11026 +/- ##
==========================================
- Coverage 92.21% 92.20% -0.02%
==========================================
Files 405 405
Lines 19240 19244 +4
==========================================
Hits 17743 17743
- Misses 1130 1134 +4
Partials 367 367 ☔ View full report in Codecov by Sentry. |
This is to address some of the memory concerns around the SDK's noop tracer provider. Instead of using the published noop tracer provider, we implement our own to avoid the cost of context propagation, which has been proven to be high enough to cause issues for end users.
Part of #10858