[jira] [Commented] (SPARK-34594) OrcColumnarBatchReader uncaught exception
[
https://issues.apache.org/jira/browse/SPARK-34594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17294681#comment-17294681
]
Dongjoon Hyun commented on SPARK-34594:
---
What do you mean the issue?
{quote}This issue occured with "spark.sql.files.ignoreCorruptFiles=true".{quote}
Apache Spark is designed to continue the jobs with
`spark.sql.files.ignoreCorruptFiles=true` in case of the corrupted
Parquet/ORC/Avro files.
If the job continues, there is no issue. It works as designed.
> OrcColumnarBatchReader uncaught exception
> -
>
> Key: SPARK-34594
> URL: https://issues.apache.org/jira/browse/SPARK-34594
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.5
>Reporter: Zhang Jianguo
>Priority: Major
> Attachments: report_01.PNG
>
>
> RecordReaderInterator#hasNext doesn't handle exception thrown from
> OrcColumnarBatchReader#nextKeyValue
> {code:java}
> // code placeholder
> override def hasNext: Boolean = {
> if (!finished && !havePair) {
> finished = !rowReader.nextKeyValue // Exception thrown from here
> if (finished) {
> // Close and release the reader here; close() will also be called when
> the task
> // completes, but for tasks that read from many files, it helps to
> release the
> // resources early.
> close()
> }
> havePair = !finished
> }
> !finished
> }
> {code}
> [https://github.com/apache/spark/blob/96f51378b598501cc5dc5791f794c8586a48b355/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/RecordReaderIterator.scala#L37]
>
> !report.PNG!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-34594) OrcColumnarBatchReader uncaught exception
[
https://issues.apache.org/jira/browse/SPARK-34594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17294269#comment-17294269
]
Zhang Jianguo commented on SPARK-34594:
---
[~dongjoon]
This issue occured with "spark.sql.files.ignoreCorruptFiles=true".
> OrcColumnarBatchReader uncaught exception
> -
>
> Key: SPARK-34594
> URL: https://issues.apache.org/jira/browse/SPARK-34594
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.5
>Reporter: Zhang Jianguo
>Priority: Major
> Attachments: report_01.PNG
>
>
> RecordReaderInterator#hasNext doesn't handle exception thrown from
> OrcColumnarBatchReader#nextKeyValue
> {code:java}
> // code placeholder
> override def hasNext: Boolean = {
> if (!finished && !havePair) {
> finished = !rowReader.nextKeyValue // Exception thrown from here
> if (finished) {
> // Close and release the reader here; close() will also be called when
> the task
> // completes, but for tasks that read from many files, it helps to
> release the
> // resources early.
> close()
> }
> havePair = !finished
> }
> !finished
> }
> {code}
> [https://github.com/apache/spark/blob/96f51378b598501cc5dc5791f794c8586a48b355/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/RecordReaderIterator.scala#L37]
>
> !report.PNG!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-34594) OrcColumnarBatchReader uncaught exception
[
https://issues.apache.org/jira/browse/SPARK-34594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17294223#comment-17294223
]
Dongjoon Hyun commented on SPARK-34594:
---
Could you try to use `spark.sql.files.ignoreCorruptFiles=true`? It's available
since Apache Spark 2.1.1.
> OrcColumnarBatchReader uncaught exception
> -
>
> Key: SPARK-34594
> URL: https://issues.apache.org/jira/browse/SPARK-34594
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.5
>Reporter: Zhang Jianguo
>Priority: Major
> Attachments: report_01.PNG
>
>
> RecordReaderInterator#hasNext doesn't handle exception thrown from
> OrcColumnarBatchReader#nextKeyValue
> {code:java}
> // code placeholder
> override def hasNext: Boolean = {
> if (!finished && !havePair) {
> finished = !rowReader.nextKeyValue // Exception thrown from here
> if (finished) {
> // Close and release the reader here; close() will also be called when
> the task
> // completes, but for tasks that read from many files, it helps to
> release the
> // resources early.
> close()
> }
> havePair = !finished
> }
> !finished
> }
> {code}
> [https://github.com/apache/spark/blob/96f51378b598501cc5dc5791f794c8586a48b355/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/RecordReaderIterator.scala#L37]
>
> !report.PNG!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-34594) OrcColumnarBatchReader uncaught exception
[
https://issues.apache.org/jira/browse/SPARK-34594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17293670#comment-17293670
]
Zhang Jianguo commented on SPARK-34594:
---
[~dongjoon]
Yeah, I have an external table. Other programs write corrupted ORC file results
in executor restart frequently.
!report_01.PNG!
> OrcColumnarBatchReader uncaught exception
> -
>
> Key: SPARK-34594
> URL: https://issues.apache.org/jira/browse/SPARK-34594
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.5
>Reporter: Zhang Jianguo
>Priority: Major
>
> RecordReaderInterator#hasNext doesn't handle exception thrown from
> OrcColumnarBatchReader#nextKeyValue
> {code:java}
> // code placeholder
> override def hasNext: Boolean = {
> if (!finished && !havePair) {
> finished = !rowReader.nextKeyValue // Exception thrown from here
> if (finished) {
> // Close and release the reader here; close() will also be called when
> the task
> // completes, but for tasks that read from many files, it helps to
> release the
> // resources early.
> close()
> }
> havePair = !finished
> }
> !finished
> }
> {code}
> [https://github.com/apache/spark/blob/96f51378b598501cc5dc5791f794c8586a48b355/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/RecordReaderIterator.scala#L37]
>
> !report.PNG!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-34594) OrcColumnarBatchReader uncaught exception
[
https://issues.apache.org/jira/browse/SPARK-34594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17293625#comment-17293625
]
Dongjoon Hyun commented on SPARK-34594:
---
Thank you for reporting, [~AlberyZJG].
- Did you hit some error there in some situation? For example, corrupted data?
- If then, does the uncaught failure causes job failure?
> OrcColumnarBatchReader uncaught exception
> -
>
> Key: SPARK-34594
> URL: https://issues.apache.org/jira/browse/SPARK-34594
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
>Affects Versions: 2.4.5
>Reporter: Zhang Jianguo
>Priority: Major
>
> RecordReaderInterator#hasNext doesn't handle exception thrown from
> OrcColumnarBatchReader#nextKeyValue
> {code:java}
> // code placeholder
> override def hasNext: Boolean = {
> if (!finished && !havePair) {
> finished = !rowReader.nextKeyValue // Exception thrown from here
> if (finished) {
> // Close and release the reader here; close() will also be called when
> the task
> // completes, but for tasks that read from many files, it helps to
> release the
> // resources early.
> close()
> }
> havePair = !finished
> }
> !finished
> }
> {code}
> [https://github.com/apache/spark/blob/96f51378b598501cc5dc5791f794c8586a48b355/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/RecordReaderIterator.scala#L37]
>
> !report.PNG!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
