How to delete data of a domo dataset

Subbu

Hello All,

 

I have a domo dataset created from a source ( Teradata ) via some tool. That domo dataset having 4 weeks of data I found that one of the four weeks data is incorrect and would like to delete only that 4th week data. I checked all options in domo I did not find an easy to delete that week data.

 

FYI, I have admin access but, still not be able to delete data from a domo dataset. One thing what I checked is to create another data set from the eixsiting dataset and leave the week where dirty data was there. But, the week data is huge which is about 130 millons of rows. Now, creating one more or derived dataset itslef is redundant and occupying the space. Do we hava any interface in dom where in I can write a simple SQL statement as below

 

delete from domodataset1 where item=1234;

 

This way I can easily delete data. 

 

 

Thanks!
Subbu.

floramc

There's a hidden function on DOMO's dataset. If you add ?_f=dataRepair to the end of a dataset url, new tab called Data Repair will pop up. In there, you can see all the uploaded data versions. If you want to clear the dataset, just delete all the versions. 

AS

Webhooks like this are mostly a backdoor for Domo employees but can be helpful for those who know them.

With data repair specifically, you can only remove pieces of the dataset that have come in as part of the historical append chain.  Datasets that replace themselves will only have one link in this append chain

AS

There isn't an interface where you can simply take a data set and delete rows.  However, like you mentioned you can use SQL Dataflows to write a statement to filter out the rows you don't want, then create a secondary data set as the output. Once you have that, then you might not need your original data set and can delete it if you want to clean things up.  That is, if your original dataset wouldn't have been updated anyway (replaced or appended to), then you can get rid of it.

 

In the works is a tool that will allow you to use Excel to load a dataset straight from Domo, make any changes you want, then publish that file back to Domo over the original data set.  But in this case 130m rows is a lot and Excel wouldn't work with that.  

IamJJ

I've run into a similar situation and essentially want to undo a MySQL manual update - did you find a way to remove unwanted records?

 

__JJ

jlazerus

Does this tool exist yet?

floramc

There's a hidden function on DOMO's dataset. If you add ?_f=dataRepair to the end of a dataset url, new tab called Data Repair will pop up. In there, you can see all the uploaded data versions. If you want to clear the dataset, just delete all the versions. 

AS

Webhooks like this are mostly a backdoor for Domo employees but can be helpful for those who know them.

With data repair specifically, you can only remove pieces of the dataset that have come in as part of the historical append chain.  Datasets that replace themselves will only have one link in this append chain

Sambo_Servicer

This was an amazing find. 
I had duplicate data appended.

I was going to blow away the entire dataset and spend days doing another incremental upload.

 

This was such a gift.

 

DomoFaye

The Data Repair feature has some limitation and does not seem to be an end-user-ready feature. If you delete a version, then you may find the dataset is no longer acceptable by Magic ETL. I reckon if you would like to remove/update some data, you can just filter and append it in Magic ETL. If you know sometimes you will need to remove/replace weekly/monthly/yearly data in the future, maybe try to partition them at the beginning by using CLI.

If we are talking about Teradata only, then I assume you were using workbench and you may be able to enable upsert functionality to update your data.

https://domohelp.domo.com/hc/en-us/articles/360043038714-Using-Upsert-in-Workbench-5-1

NotThatGreatAtExcel

I would just add for anyone else reading this.

I went to data repair, downloaded the CSV, deleted the rows I didn't want, and the replaced the CSV.

Seemed to work well