We have a environment with 5 nodes(Windows) with Elastic search version 0.90.13.
We got a issue, when we checked with "_cluster/health"
We are got the response "status:green", but number of node was displayed as 4 instead of 5.
Once we restart the Elastic Search Service, then everything was all set - we got 5 nodes with green status.
We are not sure why this issue came, and also not clear how it was get fixed post restarting the Elastic Search Service in the Node.
Request your help to understand why this issue came and how it was get fixed post restart the ES service in the server, and how to provide solution to this problem
If one of your nodes crashed at some point and you had at least one replica of all shards on that machine ES would've recreated the unavailable replicas elsewhere and made the cluster green again.
The ES cluster status was changing to "Yellow" with a pattern(5.00 pm - 6.00pm on every 4th day and make itself green around 6.30pm), can anyone suggest is there is any known issue associated with this.
We are using ES 0.90.13
Or, please confirm - like, if we mention 3 replica, then when the process was creating shards(at the time of pushing data from SQL DB to ES with NEST Wrapper) at this if I check the cluster/health then will it show "Yellow" and later once the process completed it shows "Green"
Cluster status "Yellow" can be expected temporarily when creating new indices or doing bulk-indexing. Without knowing your setup it looks like there is some external job doing this around the time you mentioned every four days. Is that possible?
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.