In all the elasticsearch examples I've seen so far, every index request that needs to use an ingest pipeline passes it as query parameter like

PUT /<target>/_doc/<_id>?pipeline=my_pipeline

Is this the only way to use the pipeline when indexing documents? Or are there also other ways to utilize the pipelines?

CodePudding user response：

That's only one of the options, there are plenty of ways to leverage ingest pipelines. You can also:

• Specify pipeline in bulk queries
• or when reindexing documents
• or when updating documents matching a query

You can also define a default pipeline to be used when indexing documents inside an index with the index.default_pipeline setting and the index.final_pipeline setting, so as not to have to specify it in each indexing request (as in your question).

You can also specify it in the elasticsearch Logstash output plugin or in the Filebeat elasticsearch output](https://www.elastic.co/guide/en/beats/filebeat/current/elasticsearch-output.html#pipeline-option-es).