Jaeger - Distributed Tracing Platform
Filtering duplicate spans #4894
-
|
I'm trying to understand the whole processing pipeline for opentelemetry and jaeger. I'm currently ingesting OTel traces using the jager-all-in-one image, just doing it like The trace is ingested properly and I can see the trace/spans in the jaeger UI just fine. Is it possible to configure jaeger in a way to either (ideally) not store the duplicate spans, or at least not display them in the UI? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
|
Our other storage backends use primary keys based on span ID, so they should naturally dedupe identical spans (although to be confirmed if that's the case for Elasticsearch). The in-memory storage does not implement such de-duplication. Overall, it seems like a reasonable function to have to dedupe identical spans on read. A bit of a caveat there is that Zipkin's legacy model assumes the same span ID between client and server, so we already had some logic to handle that. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you so much for your answer.. this indeed was the case. I switched to the |
Beta Was this translation helpful? Give feedback.
Our other storage backends use primary keys based on span ID, so they should naturally dedupe identical spans (although to be confirmed if that's the case for Elasticsearch). The in-memory storage does not implement such de-duplication.
Overall, it seems like a reasonable function to have to dedupe identical spans on read. A bit of a caveat there is that Zipkin's legacy model assumes the same span ID between client and server, so we already had some logic to handle that.