There was a rare issue when I tried to add several data nodes to my online ES cluster yesterday.
When the new data nodes joined to the cluster, it began to rebalance shards to these new data nodes.
Most of the shard relocations completed with no error, as I did before. Just one new data node has several search rejections after a shard relocation, lasted nearly 1 minute.
I have never encountered this issue before, I think shard relocation should not cause search rejection.
Is this a bug? My ES version is 7.5.2.
Do you have the full rejection error?
Is there anything in the logs of that Elasticsearch node or the master node?
I use the default log4j2.properties file and there is nothing in the log files.
It seems that shard relocation will not be recorded in the log
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.