Shard fail while rolling upgrade cluster

DevOps
,
1

I’m trying to rolling upgrade our OpenSearch cluster.
When I try rolling upgrade our cluster according to official document it made primary shard down.Document shows modifying cluster.routing.allocation.enable only at the beginning of the rolling upgrade and the end of the process.
When I followed instruction replaced node didn’t allocated shards maybe due to cluster.routing.allocation.enable: primaries
ElasticSearch rolling upgrade guide suggests re-enable shard allocation every time on node restarts.
Should I re-enable shard allocation every node replacement?

2

@seongbok It seems that OpenSearch documentation has a bug. The shard allocation must be disabled after each node restart as per ES docs.

I’ve just created a documentation pull request.

3

Thank you for your confirm. :slightly_smiling_face:

4

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.