# DataTable column to aggregate if field exists

**URL:** <https://discuss.elastic.co/t/datatable-column-to-aggregate-if-field-exists/108658>\
**Category:** Kibana\
**Created:** [November 22, 2017, 7:23am UTC](https://discuss.elastic.co/t/datatable-column-to-aggregate-if-field-exists/108658 "2017-11-22T07:23:54Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Robert\_Naccache](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/robert_naccache/32/97656_2.png) [@Robert\_Naccache](https://discuss.elastic.co/u/Robert_Naccache)\
**Post date:** [November 22, 2017, 7:23am UTC](https://discuss.elastic.co/t/datatable-column-to-aggregate-if-field-exists/108658/1 "2017-11-22T07:23:54Z")

</div>

Hello!

First time posting here, I've been looking around but haven't found anything to point me towards the right direction to solve my issue.  
I'm creating a Data Table with a Date Histogram as a bucket, and i need to display two columns.

One column is the Total amount of successful Jobs (this is a unique count of the field ['JOBID'] with ['Errorid'] being Null)  
second column is the total amount of jobs with errors (this is a unique count of the filed ['JOBID'] with ['Errorid'] having a value other than null)

I've tried using a filter but it applys to the whole table so that doesn't work for me. i also tried using a split rows with two conditions _exists_:errorid and !(_exists_:errorid) but this doesn't display it in a seperate column, rather it splits the rows which is not what i want.

my end result should look something like this:

| DateTime | Successful jobs | Failed Jobs |  
| Nov.22-2017 | 10 | 3 |  
| Nov.21-2017 | 15 | 6 |

Any insight on how to tackle this would be highly appreciated.

Thanks!

---

<div class="post-metadata">

**Author:** ![ppisljar](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/ppisljar/32/11588_2.png) [@ppisljar](https://discuss.elastic.co/u/ppisljar)\
**Post date:** [November 22, 2017, 7:44pm UTC](https://discuss.elastic.co/t/datatable-column-to-aggregate-if-field-exists/108658/2 "2017-11-22T19:44:35Z")

</div>

i think you have two options ...  
preferred if you have a lot of data and need speed: reindex your data:

- add a new field called `succesfull` which has a value 1 id ErrorId is null, 0 otherwise
- add a new field called `failed` which has value 1 if ErrorId is not null, 0 otherwise  
then you can simply sum on those fields

second way, should be ok if you don't have tons of data:

- add a scripted field to your index, with painless script like

```auto
 if (doc['ErrorId'].value === 'null') return 1;
return 0;

```

and a similar one for checking if ErrorId is not null for counting errors

- again, do a sum on this scripted fields

let me know if this solves your issue.

---

<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:** [December 20, 2017, 7:44pm UTC](https://discuss.elastic.co/t/datatable-column-to-aggregate-if-field-exists/108658/3 "2017-12-20T19:44:43Z")

</div>

Closed on December 20, 2017, 7:44pm UTC

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.