Here is the scenarios, I need help,
- I have weekly based indices.
- for months time filter, I refer last 5 indices. for example, topic_2021_1, topic_2021_2, topic_2021_3, topic_2021_4, topic_2021_5.
Suppose in the meantime of re-indexing.
- I am creating a new index, for example.
- during the re-indexing process,
some part of the data could be moving from topic_2021_1 to topic_2021_1_cut.
- the alias will be like 'topic_2021_1*'.
- so, here, when the re-indexing process is on and data has been written into topic_2021_1_cut index. the alias may read the duplicate data from these two ( topic_2021_1 & topic_2021_1_cut) indices.
Question is how the duplicates are handled. will it generate duplicate entries in the query? or how to handle this scenario of reindexing.