Hello everyone! I am having trouble with Dataprep. When I want to edit a recipe, it starts loading and gives an error "DataNotFound"Exception" and I have to collect a new sample, what takes time and money. Why does this happen continously??

[Trifacta_Alumni · Accepted solution]

Hi, Victor--

Does this happen with all of your recipes in all of your flows? You might check another flow to see if this issue also appears there. If so, you have some form of data access issue. If that is the case, check the directories listed in your User Profile to see if they are valid: https://cloud.google.com/dataprep/docs/html/User-Profile-Page_57344911

Guessing that the temp one could be the problem.

If the problem is just with this flow: the first rows sample is taken from the first set of rows from your source dataset. You may be able to identify which dataset is the problem by clicking your recipes in flow view.

If you find one that does not preview, you can try to swap out the dataset with a newly imported version of the dataset that is the problem.

If these quick checks do not work, I would suggest filing a support ticket, since this sampling is costing you money.

Cheers,

-SteveO

[4369a3091e37dd36eabf]

Good morning Steve,

Thank you for your answer.

In other flows with different data origin I don't have this problem.

I have reloaded some datasets and it has improved. One problem I detected is than the origin table had now one more column, so nothing worked. What are the other situations in which the origin dataset can stop working?

And to finish, if I reload the origin dataset all the line of recipes that starts with that dataset should work again?

Thank you for your help

[Trifacta_Alumni]

Hi, Victor--

When you say that the new version of the origin table has an additional column, what is this column? Is this table continuously updated? Is its schema (columns changed by name, order, data type, or added/deleted) changed frequently?

Schema changes to your source dataset can certainly affect whether your recipes work whenever you load from the table. Examples:

* Changed column name - any reference to that column name in your recipe is broken.

* Changed order - any range-based transformations (colA~colN) may fail to capture a moved column, which can have downstream effects.

* Changed data type - transformation steps that expect the old data type as input may fail.

* Added/deleted columns - range issues and "column not found" issues

One tip: you might apply a target schema to your recipe. This allows you to display the schema to which you are trying to wrangle in the Transformer page, which can be useful for managing changes to schema in your source.

Check it out here: https://cloud.google.com/dataprep/docs/html/Overview-of-Target-Matching_118947846

Cheers,

-SteveO

[4369a3091e37dd36eabf]

That link can be really useful. Thank you very much!