[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-21 Thread zjffdu
Github user zjffdu commented on the issue: https://github.com/apache/zeppelin/pull/2034 Thanks @prabhjyotsingh, Is it possible to keep the behavior consistent between notebook and interpreter setting page ? Because I am afraid it would confuse users when we introduce different

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-21 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 @zjffdu have made some changes to honour ZEPPELIN-1770, but this will be only when user restarts from notebook. But if a restart, edit or delete is made to an interpreter from interpreter

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-21 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 @FireArrow Yes, agreed, I too believe that there are few minimum expectation as an Admin, and as a end user that Zeppelin should work in desired way. So, if we all can come to a

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-21 Thread FireArrow
Github user FireArrow commented on the issue: https://github.com/apache/zeppelin/pull/2034 There should probably be a jira to decide the effect of restarting in different locations. For example: As an admin I expect the restart button in Interpreter to force a restart of all

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-21 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 Yes, I understand this will violate that [ZEPPELIN-1770](https://issues.apache.org/jira/browse/ZEPPELIN-1770). But right now I see following problem/challenges, and that is why I would

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-21 Thread zjffdu
Github user zjffdu commented on the issue: https://github.com/apache/zeppelin/pull/2034 @prabhjyotsingh Would this violate the behavior of [ZEPPELIN-1770](https://issues.apache.org/jira/browse/ZEPPELIN-1770) ? --- If your project is set up for it, you can reply to this email and

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-20 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 I've pushed few commits now in which restart/edit/delete of any interpreter from any where will close/terminate all the process of that interpreter. Let me know your thoughts on

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-20 Thread FireArrow
Github user FireArrow commented on the issue: https://github.com/apache/zeppelin/pull/2034 I think restart from settings in the table above should have the same behavior as edit+save. The settings page is generally considered an administrative function, and as long as there is a

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-19 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 Yes, agreed, documentation is one part, but, other part being how should Zeppelin or `RemoteInterpreterProcess` handle connection is more important. So, in the same example as

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-18 Thread zjffdu
Github user zjffdu commented on the issue: https://github.com/apache/zeppelin/pull/2034 Yeah, that seems a valid scenario. And looks like currently there's no document for these behavior, I think we need to add doc for it as well. --- If your project is set up for it, you can

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-18 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 Sure I see your point. Just one more case that I wanted to discuss; say in an enterprise world where not all users are allowed to access settings page, or let me put it this way there is a

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-18 Thread zjffdu
Github user zjffdu commented on the issue: https://github.com/apache/zeppelin/pull/2034 I think we don't need to differentiate Global/Per Note/Per User and/Per User + Impersonate. It should just close current session. And let `RemoteInterpreterProcess` to determinate whether to shut

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-18 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 Before even implemeting a solution I thought to discuss as to what I think the behaviour should be. \| Global | per note| per

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-18 Thread zjffdu
Github user zjffdu commented on the issue: https://github.com/apache/zeppelin/pull/2034 Agree with @cloverhearts, first we need to identify what's wrong in the current logic for the long term solution. --- If your project is set up for it, you can reply to this email and have your

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-18 Thread cloverhearts
Github user cloverhearts commented on the issue: https://github.com/apache/zeppelin/pull/2034 > Leemoonsoo 3 hours ago Member I think void close(String sessionId) supposed not necessarily terminate process while an interpreter process can have multiple sessions. I

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-17 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 Did a force push to re-trigger CI. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-17 Thread Leemoonsoo
Github user Leemoonsoo commented on the issue: https://github.com/apache/zeppelin/pull/2034 @prabhjyotsingh Do you mind turn on "Build pushes" option? Then CI build status will correctly displayed in this page.

[GitHub] zeppelin issue #2034: [ZEPPELIN-2133] All interpreters sometimes throw rando...

2017-02-17 Thread prabhjyotsingh
Github user prabhjyotsingh commented on the issue: https://github.com/apache/zeppelin/pull/2034 @jongyoul @Leemoonsoo @cloverhearts Please help review this. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your