aws-observability
Feature request: Kinesis exporter
Would you consider including the Kinesis exporter? This seems like a good option for people looking for a low latency buffer for telemetry before further processing it or exporting downstream.
Thanks a lot for this @nickcox! Do you have a concrete use case or is this more like a general suggestion?
Apologies @mhausenblas, I completely missed your reply.
Yes, we would like to compare latency of exporting to Kinesis (and using Firehose w/ out of the box buffering) vs the currently buggy and high latency otlp exporter.
The OTLP exporter is not the source of latency or defects and changing to the Kinesis exporter would not affect the behavior described in the linked issue. We are preparing a new release that includes https://github.com/open-telemetry/opentelemetry-lambda/pull/310/ which we expect to resolve the linked issue.
aiui, the design of the lambda collector is that it exports each batch synchronously and therefore the latency depends on the latency of the downstream system.
Our observability platform seems to have a typical latency of 250 to 1000ms. I'm pretty sure we could do better than that by exporting to Kinesis and then processing those events asynchronously. Open to other suggestions too though.
This issue is stale because it has been open 90 days with no activity. If you want to keep this issue open, please just leave a comment below and auto-close will be canceled
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
This issue is stale because it has been open 90 days with no activity. If you want to keep this issue open, please just leave a comment below and auto-close will be canceled
This issue is stale because it has been open 90 days with no activity. If you want to keep this issue open, please just leave a comment below and auto-close will be canceled
This issue was closed because it has been marked as stale for 30 days with no activity.