It sounds like if you did a Cat Indices you would find an index named "mydocs" which is why the system's preventing you from creating an alias of the same name. (You can check current alias names via Cat Aliases).
Your index aliases should be uniquely named from index names. Indices with rollover will format example INDEX_NAME-000001 (note the suffixing digit format) with INDEX_ALIAS. Here's an example walk through from our blog.
Sure I can make the rollover alias unique, however how does it know that for start it should use the mydocs index (which already contains 85Gb of data)?
Or is it not possible to create rollover on an already existing index?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.