I've a strange problem with winlogbeat. I've installed winlogbeat 1.3 on servers and receive directly on elastic port 9200 corrupted log that create indexes starting from December 2015 through now! Event if I delete all winlogbeat indexes, they recreate automatically! I've only one elasticsearch db, no cluster.
Daily indexes are automatically created based on the event @timestamp. Winlogbeat uses the event log record's TimeCreated field as the @timestamp in its events [1]. If you wish to have Winlogbeat only report events that were created within the last N hours/minutes/seconds then you can use the ignore_older settings [2]. Note that in Winlogbeat 5.x the ignore_older filter implementation has been improved and is much faster since it uses a time based query when asking Windows for event log records.
ignore_older is already set! or maybe the order is important here and having the parameter after -name: Application make it only apply for Application logs ?
Order doesn't matter. event_logs is a list of dictionaries (key/value pairs). Each - indicates a new entry in the list. Each dictionary configures an individual event log reader. In 1.3 the only options are name (required) and ignore_older (optional). In 5.x there are few more.
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.