# \_mlt ignoring the fields argument?

**URL:** <https://discuss.elastic.co/t/-mlt-ignoring-the-fields-argument/3921>\
**Category:** Elasticsearch\
**Created:** [February 14, 2011, 2:35am UTC](https://discuss.elastic.co/t/-mlt-ignoring-the-fields-argument/3921 "2011-02-14T02:35:59Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![didier\_deshommes](https://avatars.discourse-cdn.com/v4/letter/d/b9bd4f/32.png) [@didier\_deshommes](https://discuss.elastic.co/u/didier_deshommes)\
**Post date:** [February 14, 2011, 2:35am UTC](https://discuss.elastic.co/t/-mlt-ignoring-the-fields-argument/3921/1 "2011-02-14T02:35:59Z")

</div>

Hi,  
I've issued a mlt query like this:

curl -XGET '[http://localhost:9200/examinercogtree/article/4c59cc3cdbe1afb1d8004fb4/\_mlt?fields=article\_content&pmin\_word\_len=3&pretty=true](http://localhost:9200/examinercogtree/article/4c59cc3cdbe1afb1d8004fb4/_mlt?fields=article_content&pmin_word_len=3&pretty=true)'  
-d ' {"size":"1", "explain":"1" } '

In my config, `article_content` has `term_vectors` set to `yes`. When  
looking at the `explain` output from the mlt query, I notice that the  
terms used to generate the query are not solely from the  
`article_content field`, they are also from another field (which  
doesn't have term vectors enabled). Does this mean that `_mlt` is not  
respecting the `fields` argument? Also, Is there a way to ensure that  
mlt queries will fail on fields that don't have term vectors enabled?  
I understand that mlt is faster if your field has term vectors  
enabled.

didier

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [July 6, 2017, 4:12am UTC](https://discuss.elastic.co/t/-mlt-ignoring-the-fields-argument/3921/2 "2017-07-06T04:12:08Z")

</div>