Forum Discussion

Senthilnaga91's avatar
Senthilnaga91
New Contributor III
3 years ago
Solved

Scheduling Tasks Dynamically

Hello All,

I have a use case where I have to schedule my pipeline dynamically based on entries from an input table? What’s the best way to implement this? For Example: If I have a pipeline loadSalesTable, I have to look to my input scheduling table to decide what day and time to trigger the pipeline and accordingly I have to schedule it.

I don’t think we have any straightforward approach using Scheduled or Triggered task, looking for a work around. Thanks in advance.

  • Abhishek_Soni37's avatar
    Abhishek_Soni37
    3 years ago

    @Senthilnaga91 ,

    As @bojanvelevski mentioned you can use the Metadata snap pack to achieve it,

    Solution Approach:

    1. Create all tasks that you want to run dynamically as Scheduled.
    2. Read data from the source table with info on when to run a pipeline.
    3. Read the specific Task with the help of the Metadata snap pack.
    4. Assign new values and update the Task with the Metadata snap pack.

    Set values below in the pipeline parameter:

    source_proj: orgName/ProjectPath
    target_proj: orgName/ProjectPath
    taskToUpdate: Name of the task

    Sample pipeline:
    UpdateTaskInfo.slp (3.7 KB)

    Let us know if this worked.

    Cheers 🙂

5 Replies

  • bojanvelevski's avatar
    bojanvelevski
    Valued Contributor
    3 years ago

    Hi there @Senthilnaga91.

    The Snaplogic Snaps will be suitable for this use case. You have all CRUD operations at your disposal. Let me know if you still need help on this.

    Regards,
    Bojan

    • Senthilnaga91's avatar
      Senthilnaga91
      New Contributor III
      3 years ago

      I have provided more details in my previous reply. Please refer.

      • Abhishek_Soni37's avatar
        Abhishek_Soni37
        Contributor
        3 years ago

        @Senthilnaga91 ,

        As @bojanvelevski mentioned you can use the Metadata snap pack to achieve it,

        Solution Approach:

        1. Create all tasks that you want to run dynamically as Scheduled.
        2. Read data from the source table with info on when to run a pipeline.
        3. Read the specific Task with the help of the Metadata snap pack.
        4. Assign new values and update the Task with the Metadata snap pack.

        Set values below in the pipeline parameter:

        source_proj: orgName/ProjectPath
        target_proj: orgName/ProjectPath
        taskToUpdate: Name of the task

        Sample pipeline:
        UpdateTaskInfo.slp (3.7 KB)

        Let us know if this worked.

        Cheers 🙂

  • PratapKr_Gayen's avatar
    PratapKr_Gayen
    New Contributor II
    3 years ago

    I think your requirement is to trigger the a pipeline when ever there is a change in record in source table.
    For that you have to create pipeline to save the current count in SLDB and compare the count of record in current execution. This pipeline to compare and check the record count mismatch need to be schedule 15 or 30 min and if there is count mismatch, call the pipeline to load data to Sales table.

    • Senthilnaga91's avatar
      Senthilnaga91
      New Contributor III
      3 years ago

      No, let me be more precise. Our client has given a schedule to run what pipeline on which day and time (nnot a fixed schedule). It’s maintained in a RDBMS table. So, to trigger a pipeline, the process has to look into the scheduling table and trigger the pipleline on a particular day and time accordingly.

      Example: There is a record in scheduling table,
      PipeName,Date,Time
      LoadSalesTable,06-27-2023,05:00 PM CST
      LoadEmployeeTable,06-27-2023,04:00 PM CST

      So, my question is how to automate the process to trigger the pipeline ‘LoadSalesTable’ on 06-27-2023,05:00 PM CST ?

      (PS: The schedule for ‘LoadSalesTable’ might change in the future and the process should accommodate without manual intervention).