Server-client search architecture with PIT in ElasticSearch

I want to use search_after and PIT to provide consistent search results. The guide documentation suggests that PITs should be generated in the background and utilized after each search, rather than after every search.

My question here is

If I need to utilize the same PIT id every time, is it safe to expose it to the client so that the client receives the same search results every time? I want to keep the server as stateless as possible, and I want to store the PIT id somewhere, and the idea of using it directly on the server is that when the PIT id is updated (I'm thinking of generating a new PIT after an appropriate keep_alive time to keep the data fresh), certain clients might see inconsistent search results.

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