I have an application that will be feeding data to my cluster via the API. I would like to have the index rolled like a logstash feed, index-YYYY.MM.DD. I cannot figure out how to create the index for that pattern? I don't think this is a rollover. It appears rollover just uses a numeric counter for the pattern. We would like the date pattern as we will be using Curator to maintain our retention policy. Thanks for the help.
UT /logs-000001
{
"aliases": {
"logs_write": {}
}
}
Curator doesn't need to have a date stamp in the index name. You can use field_stats based ages in Curator to identify the oldest and/or youngest @timestamp field in the index, removing the need to put a timestamp in the index name.
You're much better off using a default numeric rollover naming sequence to achieve this, with the aforementioned field_stats approach in Curator. But if you must put a timestamp in your index name with a rollover approach, see this forum post.
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.