I need more insights into how reindexing with parallel slicing works like about how it allocates slices

  1. I have been using reindex api in kibana to reindex one of my index using "slice indexing with max set to 200". Out of 14 million documents (2.5 TB) of my source index, I've been able to index 10 million documents and slice id has reached the max limit of 200.[ i am doing this because asking cluster to reindex the whole thing at once used to destabilize the cluster and the process would stop in between]. I noticed in later slice ids, it has been updating more documents than creating new. Now that I have reached my max limit, how can I reindex the remaining 4 million without using "SLICE INDEXING"? Should I just simply use reindex api without any slices? or is there any faster way to reindex the remaining docs? 2. How to check if my newly reindexed index is the same exact copy of my source index(no more, no less)? Is there a way in elasticsearch to check like sourceIndex.documents == destinationIndex.documents?

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