Our Elasticsearch servers are configured with multiple nodes in a single cluster.
But for only one elastisearch node diskspace is increasing more when compared with other nodes in cluster.
We were clearing the old indices regularly to provide free disk space.
What is the root cause for it?
The Shard Legend might be helpful here. Can you look at each node in your cluster (using the Stack Monitoring UI) and check out how the shards are distributed across your nodes? I wonder if some nodes just have more shards than others
Hi @chrisronline,
I checked the Shard Legend
In our cluster out of 4 nodes only two nodes allocated with shards.
The nodes which are not allocated with shards one node is having high free diskspace,and the other node is having less free diskspace.
Do you have other services installed on the server with less free diskspace? Have you ran a disk space check on that server to figure out where the resources are used?
Did you run a disk scan on your Windows machine? I'm not very familiar with Windows system so I can't recommend anything, but I'm sure there are multiple to choose from. When you've done this, can you paste the results here?
I tried to check the disk scan on windows but couldn't find the solution.Any other instance to check why the diskspace is increasing more for one node.
See if the majority of disk space is used by a certain directory, other than Elasticsearch. If so, then you know the issue isn't related to the Elasticsearch service, and rather something else installed on that particular server.
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.