Skip to content

An example of behavior I found surprising in the Node OTEL SDK

Notifications You must be signed in to change notification settings

jhoward-ts/otel-node-sdk-timeout-example

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Running

Requires bun.

bun install

Run these in two separate shells to see the interaction. bun run ./timeout-server.ts bun run ./generator.ts

I was surprised to see timeout errors without retries: alt text And to see that the spans which failed to export / process do not seem to remain in the processor, and are immediately lost: alt text

Per the documentation here, https://github.com/open-telemetry/opentelemetry-proto/blob/main/docs/specification.md#all-other-responses, I assumed the spans would be retried, and the processor would hold them until they were successfully processed, at least until the max buffer size.

I'm not sure if this is a bug or misunderstanding on my part. Our http endpoints get transient timeouts, and instant loss is pretty detrimental to our use case. Can I change my configuration to avoid this, or will I need to implement my own batch processor or exporter?

About

An example of behavior I found surprising in the Node OTEL SDK

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published