We are using Kibana version: 4.5.0, when we try to query requiring analysis of large volume of historical data (like 6 months). Its throwing Error: Request Timeout after 30000ms.
We have 7 Elastic search servers cluster and we have kafka cluster in between logstash and Elasticsearch.
Elasticsearch version: 2.5
If we change Timeout 30000ms on kibana.yml to timeout= 90000ms will it work?
This is more of an Elasticsearch query, as it is Elasticsearch that is doing all the work behind the scenes. What is the specification of the nodes in the cluster? How much data do you have per node? How much of this are you querying when you see these very large latencies? How many indices and shards do you have in the cluster?
Node Specifications:
4 VCPUs, 320GB DISK, 16 GB RAM
index.number_of_shards: 1
Indices 652
DATA we get per day to single ES node is 2.6 gb. @Christian_Dahlqvist
How many indices do you create every day? How many of these are you querying when you see the high latency? What type of storage do you have? Have you tried to identify what is limiting performance when you are querying, e.g. CPU utilisation or iowait?
4 indices are created everyday. All 4 indices gets queried when we try to view the dashboard as we are adding visualization from all these 3 indices. We have open stack cloud environment.
That is a lot of quite small indices to query, which can be inefficient. I suspect you would be better off using weekly or monthly indices given the volumes you mentioned. As you did not answer my other questions, it it difficult to know exactly what the bottleneck is.
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.