h2o cluster shutdown abnormally during split_frame operation without any error message.

[abdullahmansoor]

H2O version, Operating System and Environment
(3.44.03, Suse SLES15SP4 Base Gold, Intel(R) Xeon(R) Gold 6336Y CPU). Not running on K8s/Hadoop.

Actual behavior
h2o cluster shutdown abruptly during split_frame function on a small dataset on the above specified machine but works fine on other machines.

Expected behavior
I want to avoid unexpected shutdown of shutdown of h2o cluster during split_frame operation. Although my dataset has 155 rows and I am splitting for 5%, resulting in 7 rows. However, I am using those 7 rows for prediction and it shouldn't fail in split_frame.

Steps to reproduce
df = h2o.import_file(dataset_file)
srRange=[0.05]
#df has 155 rows
splits = df.split_frame(ratios = srRange, seed = 1)

Upload logs
03-18 14:25:09.742 127.0.0.1:54321 23241 2931811-26 INFO water.default: GET /3/Frames/ds_small_01.hex, parms: {column_count=-1, column_offset=0, row_offset=0, full_column_count=-1, row_count=10}
03-18 14:25:09.763 127.0.0.1:54321 23241 Thread-320 INFO water.default: Orderly shutdown: Shutting down now.
03-18 14:25:09.776 127.0.0.1:54321 23241 Thread-321 INFO water.default: Orderly shutdown: Shutting down now.