We found that 7 million Task records were created in our org and we need to delete them. We were recommended to delete all tasks, but there are many we want to preserve.

We were able to create a CSV with all 7 million Id's, but attempting to delete through data loader keeps getting stuck arounds 100K records regardless of the loaded CSV's size.

Any suggestions to delete this volume of records?

  • 2
    I have personally used the data loader on millions of records without a problem, so it might be something with your environment. Have you tried updating Java and/or the Data Loader to the latest versions? Does your system have at the minimum system requirements for the data loader (at least 256MB free)? When you say it "gets stuck", do you mean it "stops responding" or simply doesn't go any further (but you can still cancel the operation)? For a situation like this, I'm afraid, the devil's in the details.
    – sfdcfox
    Commented May 19, 2016 at 0:44

1 Answer 1


According the best practices from Salesforce about When to Use Data Loader, you should use Data Loader for data between 50.000 to 5.000.000 records.

Data Loader is supported for loads of up to 5 million records.

In your case I will recommend you to try to split the csv into 2 or 3 files and try again, if even with this you are still having issues, check the last Data loader version (like sfdcfox said).

If data loader find any error it will create another csv with the description of the error, with this you can see what is your real problem.

I hope it helps...

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .