ETLBatch Pipeline with S3 Batch Source takes a long time to launch the MapReduce Job

Description

The problem seems to be caused due to the unusually long delay during the computation of input splits. Because of this the MR job is launched one hour after the workflow application is started. Thread dump is attached.

Release Notes

None

Attachments

1

Activity

Show:
Won't Fix
Pinned fields
Click on the next to a field label to start pinning.

Details

Assignee

Reporter

Affects versions

Priority

Created December 11, 2015 at 6:29 AM
Updated June 1, 2020 at 9:22 PM
Resolved June 1, 2020 at 9:22 PM