I have two clusters, one running 2.4 and the other 5.2.2. I created an S3-repo on the 2.4 cluster, and took a snapshot to s3. I then added this repo on the 5.2.2 cluster, and restored to it succesfully. Now, we have some more indices that need to be transferred over and some that need to be updated. I deleted the ones to be updated on the new cluster, and took a new snapshot on the 2.4 cluster, with that same repository.
The new (I tried a second time) snapshots show up fine in the 2.4 cluster, however neither of them show up in the 5.2.2 cluster, it still only shows the original one. I've confirmed the files are in the s3 bucket.
Oh, the documentation didn't mention that at all, it just says you can restore a snapshot from 2.x onto a 5.x cluster, so that was how I thought of doing it. Is there a different method you would recommend, or would simply removing the repo from the 2.x cluster, then adding it to 5.x work?
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.