Hi there,
I hope someone can help me with a very specific problem I have.
We integrated Elasticsearch in our multi-tenant-system.
Based on our system we forced to use update_by_query
instead of update
.
This system creates multiple consecutive updates to Elasticsearch for the same document in under 1 second.
Because of this we will lose some data from requests because some updates are not executed.
I guess that the data is lost becuase a document is available for new updates in 1 second.
Now my question: Is there any possibility that these request are not lost? Is there some kind of cache that can be activated that I overlooked in the documentation?