Question about Opaque ID in index requests and Opaque ID restrictions/limitations

Hello Team,

We are currently adding Opaque ID to our Elasticsearch calls to improve traceability in our system. We have also enabled slow logs for both indexing and searching, where Opaque ID is very helpful.

So far, I have checked that X-Opaque-Id header of all of my requests (search, index, bulk, scroll, update ...) reached the Elasticsearch server and also got back in response header.

For some requests, Opaque ID is also added to log events in search slow logs (search, scroll requests).

But for some requests, Opaque ID is not added to the log events in slow logs: at least for index request and bulk request containing index action. This is probably due to the structure of the log event where 'id' field is already used for document id.

I have two questions regarding using Opaque ID in Elasticsearch requests:

  1. Is there any other info about using Opaque ID in index requests or should I open an bug/enhancement Issue in Elasticsearch GitHub repository?

  2. Are there any other restrictions or limitations on using Opaque ID (across all requests) which I am not aware of? For example,using Opaque ID in bulk requests etc. ?

Thank you very much for the info.

We are using Elasticsearch 7.17.9
(preparing for upgrade to Elasticsearch 8.6)

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.