Vira Vitanska created DATALAB-2433:
--------------------------------------
Summary: [GCP]: Investigate performance Dataproc and libaries usage
Key: DATALAB-2433
URL: https://issues.apache.org/jira/browse/DATALAB-2433
Project: Apache DataLab
Issue Type: Task
Security Level: Public (Regular Issues)
Components: DataLab Main
Reporter: Vira Vitanska
Assignee: Mykola Bodnar
Fix For: v.2.5
Attachments: Playbook which was running previuosly successfully.png,
Reshape2 installation.png, ggplo2 Jupyter UI.png, ggplot installation via
Jupyter UI.txt, ggplot2 ssh.png
*Preconditions:*
# Dataproc based on CPU is created on Jupyter
# Dataproc Master/slave are n1-highmem-4
*Case1:* Flights data Visualization R playbook for Dataproc
v.2.0.0-RC22-ubuntu18 requires libraries from R package ggplot2 and reshape2.
ggplot 2 was installed on Dataproc master/slaves via ssh successfully. However
running Flights data Visualization R playbook does not find ggplot2.
If ggplot2 install via Jupyter UI running Flights data Visualization R
playbook does find library ggplot2.
*Questions:*
# Is the library installed in different places?
# Why is not found the library if install via ssh terminal Dataproc?
*Case2:* Reshape2 installation was more than 40 minutes via Jupyter UI and
after that i restart kernel. The attempt of reshape2 installation or running
previously successful playbook causes a new error:
{code:java}
The code failed because of a fatal error:
Invalid status code '500' from http://172.31.16.13:8998/sessions with
error payload: "java.lang.NullPointerException".
Some things to try:
a) Make sure Spark has enough available resources for Jupyter to create a Spark
context.
b) Contact your Jupyter administrator to make sure the Spark magics library is
configured correctly.
c) Restart the kernel.
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
