As we know elasticsearch stores the logstash indices in the format logstash-yyyy.mm.dd. Does elastic search creates new indexes for new date by re-indexing previous day indexes?
Ex: I am observing every day logstash-* folders content in getting increased by twice the previous day size.
logstash-2015.11.17 was ~500MB,
logstash-2015.11.18 was ~1.5 GB and
logstash-2015.11.19 is > 3 GB
As we know elasticsearch stores the logstash indices in the format logstash-yyyy.mm.dd. Does Elasticsearch creates new indexes for new date by re-indexing previous day indexes?
No, Elasticsearch doesn't reindex data on its own.
Ex: I am observing every day logstash-* folders content in getting increased by twice the previous day size.
logstash-2015.11.17 was ~500MB,
logstash-2015.11.18 was ~1.5 GB and
logstash-2015.11.19 is > 3 GB
With respect to above query below is some more information/issues.
As per Marvel
logstash-2015.11.19 data size is 1,004.8MB and Document Count is 2.4m
logstash-2015.11.20 data size is 263.7MB and Document count is 618.5k
But I am not able to view any data related to logstash-2015.11.20 in Kibana.
Later investigated the log files and got to know last update to elasticsearch index happened on 2015-11-19 22:04:09,842 and last good contact between logstash and elasticsearch is at ~Thu Nov 19 23:00:00 CST 2015.
So while creating new indices folder by elasticsearch, logstash is loosing connectivity with elasticsearch. I have observed the same behaviour on 2015.11.17 but restarting logstash instance resolved the problem but triggered my initial query on this topic.
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.