Tracing of operations #18
Description
Community Note
- Please vote on this issue by adding a 👍 reaction to the original issue to help us know what to work on
- Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
Description
Currently we do monitoring using metrics (histograms), but some users want to have more precision and answer to questions like: which arguments caused an operation to be slow.
For that we need tracing. In general we consider that tracing should be used when need be and not necessarily on all operations all the time.
Steps
- Build an API to accept traces
- Add tracing to the SDKs
- Add configuration using Remote configuration #14