Neetu Gupta created HUDI-3915:
---------------------------------
Summary: Error upserting bucketType UPDATE for partition :0
Key: HUDI-3915
URL: https://issues.apache.org/jira/browse/HUDI-3915
Project: Apache Hudi
Issue Type: Bug
Components: deltastreamer
Reporter: Neetu Gupta
I have updated the hudi column partition from 'year,month' to 'year. Then I ran
the process in overwrite mode. The process executed successfully and hudi table
got created.
However, when the process got triggered in 'append' mode, I started getting the
error mentioned below:
'
Task 0 in stage 32.0 failed 4 times; aborting job java.lang.Exception: Job
aborted due to stage failure: Task 0 in stage 32.0 failed 4 times, most recent
failure: Lost task 0.3 in stage 32.0 (TID 1207, ip-10-73-110-184.ec2.internal,
executor 6): org.apache.hudi.exception.HoodieUpsertException: Error upserting
bucketType UPDATE for partition :0 at
org.apache.hudi.table.action.commit.BaseSparkCommitActionExecutor.handleUpsertPartition(BaseSparkCommitActionExecutor.java:305)
'
Then I reverted the partition columns back to 'year,month' but still got the
same error. But, when I am writing data in different folder in 'append' mode,
the script ran fine and I could see the Hudi table.
In short, the process is not working when I am trying to append the data in the
same path. Can you please look into this. This is critical to us because the
jobs are stuck.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
