You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@spark.apache.org by charles li <ch...@gmail.com> on 2016/02/01 03:43:37 UTC
confusing about start ipython notebook with spark between 1.3.x and 1.6.x
I used to use spark 1.3.x before, and explore my data in ipython [3.2]
notebook, which was very stable. but I came across an error
" Java gateway process exited before sending the driver its port number "
my code is as bellow:
```
import pyspark
from pyspark import SparkConf
sc_conf = SparkConf() ### error occurs here
```
then I ask google for help, here is a answer on stackoverflow:
http://stackoverflow.com/questions/30763951/spark-context-sc-not-defined/30851037#30851037
, it says:
```
One solution is adding pyspark-shell to the shell environment variable
PYSPARK_SUBMIT_ARGS:
export PYSPARK_SUBMIT_ARGS="--master local[2] pyspark-shell"
There is a change in python/pyspark/java_gateway.py
<https://github.com/apache/spark/commit/e3f315ac358dfe4f5b9705c3eac76e8b1e24f82a#diff-d88dad45f458b7ad9f1c6a63b07e3a28>,
which requires PYSPARK_SUBMIT_ARGS includes pyspark-shell if a
PYSPARK_SUBMIT_ARGS variable is set by a user.
```
then I change my PYSPARK_SUBMIT_ARGS from `--master spark://
10.21.208.21:7077 --deploy-mode client` to `--master spark://
10.21.208.21:7077 --deploy-mode client pyspark-shell` , it does works, but
it raise another question, each time when I create sc in different
notebooks, the spark app name is `pyspark-shell` even though I explicitily
set the app name using SparkConf, that's really confused me these days.
then My questions come:
- How to start ipython notebook with spark integrated in spark 1.6.0;
- why it works when I set `pyspark-shell` in PYSPARK_SUBMIT_ARGS when
start ipython notebook with spark 1.6.0;
- why it does not work when I explicitly set the app name using
SparkConf;
great thanks.
--
*--------------------------------------*
a spark lover, a quant, a developer and a good man.
http://github.com/litaotao