Skip to content

Conversation

@breedx-splk
Copy link
Contributor

Fixes #2257.

I don't love this, but it's a way for us to work around this empty string Attribute issue until we can get some clarification around the spec. See open-telemetry/opentelemetry-specification#4660.

@otelbot-java-contrib
Copy link
Contributor

🔧 The result from spotlessApply was committed to the PR branch.

Copy link
Contributor

@LikeTheSalad LikeTheSalad left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you for taking a look into this. If I understand correctly, the issue appeared after disk-buffering swapped its serialization mechanism to use the core SDK serialization. Could it mean that there's an issue with the upstream serialization that we should address (regardless of the otel spec changes, as it seems more related to proto spec issues)?

@breedx-splk
Copy link
Contributor Author

If I understand correctly, the issue appeared after disk-buffering swapped its serialization mechanism to use the core SDK serialization.

Right yeah see this comment open-telemetry/opentelemetry-java#7656 (comment)

Could it mean that there's an issue with the upstream serialization that we should address (regardless of the otel spec changes, as it seems more related to proto spec issues)?

That was my instinct as well...but I started to pull that thread and the sweater got very ugly very quickly. See open-telemetry/opentelemetry-java#7656. Ideally yes, it should be handled in upstream....so if you're hesitant to accept this on ideological grounds, I understand...but I wanted to get a temporary fix in anyway.

Copy link
Contributor

@LikeTheSalad LikeTheSalad left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That was my instinct as well...but I started to pull that thread and the sweater got very ugly very quickly. See open-telemetry/opentelemetry-java#7656. Ideally yes, it should be handled in upstream....so if you're hesitant to accept this on ideological grounds, I understand...but I wanted to get a temporary fix in anyway.

Got it. It does look like a tricky issue to tackle, I agree it's probably better/faster to have some workaround here in the meantime. Thank you.

@breedx-splk breedx-splk added this pull request to the merge queue Sep 18, 2025
Merged via the queue into open-telemetry:main with commit c2c62aa Sep 18, 2025
25 checks passed
@breedx-splk breedx-splk deleted the empty_attribute_disk_buffering branch September 18, 2025 20:32
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

DiskBuffering crashes with empty string attributes

3 participants