ADF dataflow is creating a 0 Byte file only when Source is configured to move file after completion

Aditya Verma 20 Reputation points
2024-04-22T23:02:45.4433333+00:00

ADF provides an option to move the source file after dataflow completion by setting up options in 'Source Settings'. We're trying to use this option to move our source file from SFTP server to an archive location (another folder on same SFTP server)

From: SFTP/source/1.csv
To: SFTP/source/archive/1.csv

User's image

The pipeline/dataflow completes without an error, but it doesn't move the source file to archive, instead it just creates a 0Bytes file in Archive folder with same name.

Please advise how can this be achieved at Dataflow level, without using Pipeline Copy activity?

Azure Data Factory
Azure Data Factory
An Azure service for ingesting, preparing, and transforming data at scale.
10,197 questions
0 comments No comments
{count} votes

Accepted answer
  1. AnnuKumari-MSFT 32,161 Reputation points Microsoft Employee
    2024-05-15T08:22:00.8766667+00:00

    @Aditya Verma ,

    Thankyou for sharing the details about resolution of your query. It would certainly help others experiencing the same thing in future and can easily reference this! Since the Microsoft Q&A community has a policy that "The question author cannot accept their own answer. They can only accept answers by others ", I'll repost your solution in case you'd like to "Accept " the answer.

    'From' and 'To' path options expects the path to be pointing to the folderpath and not the filename. Removing the filename from the 'From Path' option helped in resolving the issue.

    For example:
    From: FOLDER/SUBFOLDER
    To: FOLDER/SUBFOLDER/ARCHIVE
    
    

    This functionality also works well for the Wildcard file mode option. ADF identifies the file being processed and moves it from the 'FROM' to the 'TO' directory specified.

    Kindly consider accepting the answer by clicking on Accept answer button. Thankyou

    0 comments No comments

4 additional answers

Sort by: Most helpful
  1. AnnuKumari-MSFT 32,161 Reputation points Microsoft Employee
    2024-04-23T08:42:13.95+00:00

    Hi Aditya Verma ,

    Welcome to Microsoft Q&A platform and thanks for posting your query here.

    As per my understanding you are trying to use the "After completion" feature to move the source file after successful copy of the data in mapping dataflow. However, it's not working as expected in your case.

    I tried to repro your scenario where I am trying to move the source file from one folder in ADLS container to a subfolder after successful competition of data copy and it is working as expected, unlike your case .

    My observation is that we do not need to mention the file name in the 'to path ' , we just need to mention the folder/subfolder path without appending filename.

    Also, kindly validate if other transformations till sink transformation ran successfully or not , and sink file got created without any issue or not. After its successful copy to sink, source file should be moved with data.

    Kindly share the screenshot of entire dataflow configurations including source and sink settings. Thankyou

    Hope it helps. If it resolved your issue, kindly hit on Accept answer button else kindly share more details for better help. Thankyou

    0 comments No comments

  2. Aditya Verma 20 Reputation points
    2024-04-23T14:05:20.3433333+00:00

    It is a very simple dataflow which reads from SFTP host and writes to Azure SQL DB.

    The data gets populated as expected in the sink db table, but source file doesn't gets moved after completion.

    User's image

    Please see source parameters as below:

    User's image

    User's image

    Below is the behavior based on two options:

    1. filename provided alongwith archive path in 'TO' column: Pipeline completes successfully with no errors and 0 Bytes file gets generated in archive path but source file is not deleted.
    2. only Archive path provided in 'TO' column: Pipeline completes successfully with no errors but no file is generated in archive path directory and source file is also not deleted.
    0 comments No comments

  3. Aditya Verma 20 Reputation points
    2024-04-23T17:56:36.62+00:00

    Following up on this question, I realized that its not the 0 Bytes folder, instead a folder gets created with the filename. Below is updated comments

    Below is the behavior based on two options:

    1. filename provided alongwith archive path in 'TO' column TO: FOLDER/SUBFOLDER/FILE
      :Pipeline completes successfully with no errors and empty folder gets generated in archive path but source file is not moved.
    2. only Archive path provided in 'TO' column TO: FOLDER/SUBFOLDER
      Pipeline completes successfully with no errors but no file is generated in archive path directory and source file is also not deleted.
    0 comments No comments

  4. Aditya Verma 20 Reputation points
    2024-04-24T03:09:53.81+00:00

    I tried few more options, and ultimately got it working by omitting the filename from 'From' option.

    I.e.

    From: FOLDER/SUBFOLDER

    To: FOLDER/SUBFOLDER/ARCHIVE

    This even worked well for Wildcard file mode option. ADF identifies the file being processed and moves from 'FROM' to 'TO' directory listed.

    User's image

    0 comments No comments