Job running FAQ

What do I do when the download button does not respond or loading is interrupted during approval-free result download in the MaxCompute console?

Problem

When you download query results without approval in the MaxCompute console, the download button does not respond or loading is interrupted.

Cause

Browser cache or installed plugins may interfere with the download process.

Solution

Try the following methods in order:

1. Open Chrome in incognito mode, re-run the query, and then perform the download. Incognito mode disables extensions and ignores cached data, which helps rule out interference from browser cache or plugins.

2. If the issue persists in incognito mode, clear your browser cache and cookies, and then retry the download in a normal browser window.

3. If the issue still persists, try a different browser.

If none of the preceding methods resolve the issue, press F12 to open the browser developer tools, check the Console tab for error messages, and submit a ticket with the error details.

After the download succeeds, the file is saved to your local download directory.

Why does a scheduled task fail but succeed when manually re-run?

Problem

A scheduled MaxCompute task fails during automatic scheduling but succeeds when you manually re-run it.

Cause

This issue is typically caused by source data fluctuation. For example, during early morning hours, the source (such as Hologres) may write dirty data that contains overly long fields. This triggers a buffer overflow error when MaxCompute reads the data. After the dirty data is updated or cleaned, the manual re-run reads normal data and succeeds.

Solution

To identify and resolve the issue:

1. Check the data quality and historical write records of the source. Compare the timestamps of the dirty data writes with the scheduled task failure times to confirm the correlation.

2. Examine the source data for records with overly long fields or other anomalies that could trigger buffer overflow errors.

3. Clean or correct the identified dirty data in the source.

After you clean the dirty data, re-run the scheduled task. If the task completes without errors, the issue is resolved.

How do I troubleshoot the DATAX_R_ODPS_006 error in DataWorks data synchronization tasks?

Problem

A DataWorks data synchronization task fails with the error DATAX_R_ODPS_006, which indicates that the AccessKey cannot be obtained.

Cause

This error typically occurs when the task execution context becomes invalid. Common causes include changes to the datasource config, a disabled task owner account, or insufficient permissions for the current logon account.

Solution

Troubleshoot the issue by following these steps:

1. Check the task history for successful records and identify when the failure started. If the task started failing after a specific point in time, check for configuration or account changes made around that time.

2. Check whether the source and destination datasource config has been changed. If the configuration was changed, verify that the connection information, including the AccessKey, is correct. Alternatively, restore the original configuration.

3. Verify whether the task owner has been changed. If the original owner's account is disabled, assign a new owner who has the required permissions.

4. Confirm the type of the current logon account, such as a Resource Access Management (RAM) user or role. Ensure that the account has the necessary permissions to access the data sources.

5. After you change the task owner, resubmit and publish the task so that the new execution context takes effect.

After you complete the preceding steps, run the task again. If the task runs without the DATAX_R_ODPS_006 error, the issue is resolved. If the error persists, submit a ticket and include the task ID, error details, and the troubleshooting steps that you have already taken.