andygrove opened a new issue, #498:
URL: https://github.com/apache/datafusion-comet/issues/498
### Describe the bug
It looks like we may have a hard-coded assumption in shuffle where we cannot
process an incoming batch that is larger than the default batch size.
I have seen this error more than once but currently have a repro so will
document this more as I investigate.
```
org.apache.comet.CometNativeException: range end index 8221 out of range for
slice of length 8192
...
at org.apache.comet.Native.executePlan(Native Method)
at
org.apache.comet.CometExecIterator.executeNative(CometExecIterator.scala:72)
at
org.apache.comet.CometExecIterator.getNextBatch(CometExecIterator.scala:129)
at
org.apache.comet.CometExecIterator.hasNext(CometExecIterator.scala:144)
at
org.apache.spark.sql.comet.execution.shuffle.CometShuffleWriteProcessor.write(CometShuffleExchangeExec.scala:492)
at
org.apache.spark.sql.comet.shims.ShimCometShuffleWriteProcessor.write(ShimCometShuffleWriteProcessor.scala:35)
at
org.apache.spark.sql.comet.shims.ShimCometShuffleWriteProcessor.write$(ShimCometShuffleWriteProcessor.scala:28)
at
org.apache.spark.sql.comet.execution.shuffle.CometShuffleWriteProcessor.write(CometShuffleExchangeExec.scala:448)
```
### Steps to reproduce
_No response_
### Expected behavior
_No response_
### Additional context
_No response_
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
