Hi Sachin Janani, There're no special reason that interpreter runs only on the localhost, except for current interpreter starting mechanism support only localhost.
So there should be code modification. RemoteInterpreterProcess.java[1] and bin/interpreter.sh[2] are directly related codes. Also you might want to check this proposal [3]. Thanks, moon [1] https://github.com/apache/incubator-zeppelin/blob/master/zeppelin-interpreter/src/main/java/org/apache/zeppelin/interpreter/remote/RemoteInterpreterProcess.java [2] https://github.com/apache/incubator-zeppelin/blob/master/bin/interpreter.sh [3] https://cwiki.apache.org/confluence/display/ZEPPELIN/Cluster+Manager+Proposal On Tue, May 31, 2016 at 10:10 AM Sachin Janani <sjan...@snappydata.io> wrote: > Hi, > I am writing a custom interpreter for one of the service for which I > want the remote interpreter to be started on some remote machine > (other than localhost). Also we have a usecase wherein our service > will start interpreter before zeppelin server.But after looking to the > zeppelin code I found that RemoteInterpreterProcess.java starts > interpreter only on the localhost i.e the machine on which zeppelin > server is running and also there I didnt found anyway where zeppelin > can connect to the RemoteInterpreter instead of starting a new one. > My question here is, > a) was there any specific reason because of which interpreter was > started on localhost? > b) Is there any way we can allow zeppelin server to connect to already > running remote interpreter ? > > Thanks, > Sachin Janani >
