mtsol issues

Results 5 issues of


                                            mtsol

mpirun failed with exit status 1 while running spark submit with master yarn on horovod

**Environment:** 1. Framework: (Keras) 2. Framework version: 2.6.0 3. Horovod version: 0.25.0 4. MPI version: 3 5. CUDA version: 11.7 6. NCCL version: NA 7. Python version: 3.8 8. Spark...

bug

spark

spark (master yarn fail) PMIX ERROR: OUT-OF-RESOURCE in file dstore/pmix_esh.c at line 831 An error occurred while calling z:org.apache.spark.api.python.PythonRDD.collectAndServe. : org.apache.spark.SparkException: Job 4 cancelled part of cancelled job group horovod.spark.run.0 RuntimeError: mpirun failed with exit code 1

I am using horovod docker provided environment not any custom made. **Environment:** 1. Framework: (pytorch, tensorflow, Keras) 2. Framework version: 2.6 3. Horovod version: 0.25.0 4. MPI version: 3.0.0 5....

bug

spark

will ucx release its support for spark >= 3.1

java.lang.VerifyError: class org.apache.spark.shuffle.UcxShuffleManager overrides final method getReader.(Lorg/apache/spark/shuffle/ShuffleHandle;IILorg/apache/spark/TaskContext;Lorg/apache/spark/shuffle/ShuffleReadMetricsReporter;)Lorg/apache/spark/shuffle/ShuffleReader;

Getting this error

Very old versions of all libraries

Why you have not updated any library in your framework? it is not supporting pagination?