You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When attempting to start spark with RAPIDS, I consistently face issues where the executors lose connection, indicated by the message "command exited with code 1," followed by reconnection attempts, which are repeatedly made.
24/06/07 13:13:21 ERROR TaskSchedulerImpl: Lost executor 0 on {IP}: Command exited with code 1
24/06/07 13:13:23 ERROR TaskSchedulerImpl: Lost executor 1 on {IP}: Command exited with code 1
Additionally, when using spark-submit, I encounter a warning regarding GPU overrides:
24/06/07 13:15:29 WARN GpuOverrides:
! <RowDataSourceScanExec> cannot run on GPU because GPU does not currently support the operator class org.apache.spark.sql.execution.RowDataSourceScanExec
@Expression <AttributeReference> min#0 could run on GPU
@Expression <AttributeReference> max#1 could run on GPU
This warning is followed by further losses of executors.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
When attempting to start spark with RAPIDS, I consistently face issues where the executors lose connection, indicated by the message "command exited with code 1," followed by reconnection attempts, which are repeatedly made.
Additionally, when using spark-submit, I encounter a warning regarding GPU overrides:
This warning is followed by further losses of executors.
nvidia-smi:
versions:
Beta Was this translation helpful? Give feedback.
All reactions