i am using livy interpreter, i don't have any issues now.
it's bit confusing whole impersonation on interpreter level and then on the
spark yarn level.
we should have some good documentation explaining this.
how can i contribute to documentation?
On Tue, Aug 15, 2017 at 8:49 AM, Arun Natva wrote
If you are using zeppelin 0.7.0, the easiest way is to use Livy interpreter and
get rid of the zeppelin impersonate CMD confit from zeppelin-env
Livy interpreter supports user impersonation. You can use Livy.spark,
livy.pyspark, livy.sql interpreters
Sent from my iPhone
> On Aug 15, 2017, at 9
The impersonation feature of Zeppelin is currently based per default on
SSH-ing to localhost:
https://zeppelin.apache.org/docs/0.7.0/manual/userimpersonation.html
This can be overridden e.g. by setting ZEPPELIN_IMPERSONATE_CMD as env
variable (e.g. for sudo), but the above error you are reporti
Great!
On Wed, May 10, 2017 at 5:29 AM, Yeshwanth Jagini
wrote:
> i am not able to recreate this problem.
>
> it is working now
>
> On Tue, May 9, 2017 at 1:22 PM, Jongyoul Lee wrote:
>
>> It looks like a ssh problem. Can you tell me the setting how we use Spark
>> Interpreter?
>>
>> On Fri, Ma
i am not able to recreate this problem.
it is working now
On Tue, May 9, 2017 at 1:22 PM, Jongyoul Lee wrote:
> It looks like a ssh problem. Can you tell me the setting how we use Spark
> Interpreter?
>
> On Fri, May 5, 2017 at 7:09 AM, Yeshwanth Jagini
> wrote:
>
>> Any idea why i am running
It looks like a ssh problem. Can you tell me the setting how we use Spark
Interpreter?
On Fri, May 5, 2017 at 7:09 AM, Yeshwanth Jagini
wrote:
> Any idea why i am running into this issue,
>
> INFO [2017-05-04 22:02:24,996] ({pool-2-thread-2}
> SchedulerFactory.java[jobFinished]:137)
> - Job par
