[IPython-dev] From Pyro to IPython/Jupyter

[Frédéric Mantegazza]

Le 02/12/2015, Thomas a écrit :

> Note that our machinery doesn't make this very easy at present - e.g. the
> notebook server only talks to kernels it started, so you can't create a
> notebook and attach it to an existing kernel. And running a console or a
> qt console with a kernel on another machine involves setting up SSH
> tunnels.

Ok. SSH is fine for consoles.

> One simple thing you could do would be to run the notebook server itself
> on the machine controlling the instrument - possibly under Jupyterhub, so
> users have separate logins. But that's obviously quite a large service to
> have running on a computer directly connected to important hardware.

Fortunately, people here don't have bad intentions :o) For example,
login/password strategy on instruments machines is very simple, and almost
anyone can log in any machine. As far as I know, nobody has ever messed up
somebody else experiment ;o)

And all stuff is protected from the outside by strong firewall.

Jupyterhub can be interesting to give a larger access to the instrument
responsible, and simple access to standard users (I understand that this
is something we will have to implement, and not provided by Jupyter).

-- 
    Frédéric MANTEGAZZA             CEA-Grenoble
    Tel.     : 33 (0) 476 207 617   INAC/SPSMS/MDN
    Fax      : 33 (0) 476 483 906   17, rue des Martyrs
    Courriel : mantegazza at ill.fr    F-38054 Grenoble Cedex 09