[ https://issues.apache.org/jira/browse/METRON-1829?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Justin Leet updated METRON-1829: -------------------------------- Fix Version/s: 0.7.0 > Large Error Message Causes Slow Search Performance > -------------------------------------------------- > > Key: METRON-1829 > URL: https://issues.apache.org/jira/browse/METRON-1829 > Project: Metron > Issue Type: Bug > Reporter: Ryan Merriman > Assignee: Ryan Merriman > Priority: Major > Fix For: 0.7.0 > > > Errors that occur during batch writes in the index topologies (batch and RA) > are written to Elasticsearch as a single, large error message, with a field > for each failed message. For example, if the batch size is 5000, a single > error message will be created with 5000 fields `raw_message_0`, > `raw_message_1`, .., `raw_message_4999`. With such large messages, searching > the error index in Elasticsearch is excessively slow. -- This message was sent by Atlassian JIRA (v7.6.3#76005)