I have four nodes in cluster .All my 4 nodes are master and data enabled. let say node1 is primary master .I am indexing data to ES cluster so primary master will take responsibility of indexing incoming data right?
so primary master will decide where to index the incoming data?.let say this doc is pushing to node4 at that time we have some gc pausing is going on on node4.so node 4 is not reachable by primary master(node1) at this time what primary master will do.will it wait to index requested doc to node 4 or it will push to any other nodes in cluster or if the request is timed out it wont index that particular data?
Thank you for the reply. let say if primary master is calculating that routing sequence to node 4 to index document in this case if request is timed out we will lose data right?please correct me if i am wrong..since es is a cluster is there any mechanism of node 4 not reaching to primary master ,primary master will recalculate routing details and index to some other shard in the same index.
The response you get back from ES will state it couldn't process it, so as long as you are handling those response codes adequately then you won't lose anything.
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.