Sparkling water hanging during as_h2o_frame

Description

Hello,

I found a hanging issue with h2o_pysparkling_2.2-2.2.16

I have attached the test.py script, no logs as it's easily reproducible.

spark-submit --master yarn test.py

The code hangs during the spark_to_h2o conversion:
`h2o_frame = hc.getOrCreate(spark).as_h2o_frame(my_df)`

However, when checking H2O flow, the H2OFrame is succesfully created, but the code doesn't go on (from the little research I did on it, I was able to see that a GET was not correctly being sent)

I ended up resolving our issue in production by downgrading to h2o_pysparkling_2.2-2.2.10, though, admittedly, I didn't try any of the versions inbetween.

More info:
sparkling water jar: h2odriver-sw2.2.10-hdp2.6-extended.jar
spark: 2.2.0.2.6.3.0-235
hadoop: 2.6.3
H2O: I've encountered this with 3.18.0.4 and 3.18.0.10

Assignee

Jakub Hava

Reporter

Daniel Neagoe

Labels

CustomerVisible

No

testcase 1

None

testcase 2

None

testcase 3

None

h2ostream link

None

Affected Spark version

None

AffectedContact

None

AffectedCustomers

None

AffectedPilots

None

AffectedOpenSource

None

Support Assessment

None

Customer Request Type

None

Support ticket URL

None

End date

None

Baseline start date

None

Baseline end date

None

Task progress

None

Task mode

None

Priority

Blocker
Configure