memory leaks reported by valgrind when otlp http exporter is used #3641
Unanswered
farzinlize
asked this question in
Q&A
Replies: 1 comment 1 reply
-
^^^ these are all objects with static lifetime (like file static/global objects). |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Even though I call
tracerProvider->shutdown()
at the end of my app, valgrind reports many memory leaks related to otlp exporter (libgnutlsm, google protobuf and such). There is no leak when using the same app with OStreamSpanExporter instead. For cleanup, I set an empty TracerProvider just like examples and before that, I calledprovider->ForceFlush
andprovider->Shutdown
using the exact pointer I created when setting up the TracerProvider in initial phase. it complies and in runtime I see no errors. I also confirmed that shutdown method and forceFlush method returning 1(true) so I guessed it means no error.please guide me how to implement a good cleanup function for OpenTelemetryC++ library so no error and memory leak problem appears on valgrind report. (I don't want to suppress such errors in valgrind report, ChatGPT already suggested that)
here is some example errors from valgrind report:
Beta Was this translation helpful? Give feedback.
All reactions