Hi, we are running a 7 machine cluster in EC2 and when were expirencing a
little inconsistency when we run the same query against a different machine.
We have a load balancer in front so each request is handled by a different
node, however we are getting about 5% mismatch with the facets.
We have tried flushing the index, and clearing the index cache but none of
these help and the facet counts still change from node to node.
I have already seen this some weeks ago.
One of my node had inconsistent data. I stopped the node, let the shards been replicated in the cluster, delete data dir and restart ES.
But, if you are unlucky, you will perhaps have to reindex...
HTH
--
David
Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs
Hi, we are running a 7 machine cluster in EC2 and when were expirencing a little inconsistency when we run the same query against a different machine.
We have a load balancer in front so each request is handled by a different node, however we are getting about 5% mismatch with the facets.
We have tried flushing the index, and clearing the index cache but none of these help and the facet counts still change from node to node.
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.