[IPython-dev] Unable to start Jupityr on Docker using "Quick start" instructions

[Kyle Kelley]

Hey Robert!

Couple things to get started in diagnosing these issues with docker.

Use docker ps to list running containers and docker ps -a to list all
containers including those that have stopped. If any jupyter/demo servers
or jupyter/tmpnb instances were up, use their container id (e.g. 8239ba87)
to run docker logs [CONTAINER].

That being said, the big change that was introduced in the last week was
spawn pools. It now takes longer for tmpnb to boot while it is allocating
containers. That *does* mean that you won't see the service up immediately.
Check the logs to see what's going on. You also need to make real choices
about memory limits on each container (--mem-limit=512m) as well as pool
size (--pool-size=10). pool_size*mem_limit should be < available RAM.

I deployed this earlier tonight with the vanilla docker instructions, but I
gave it some time to warm up.

If none of this helps, I can work with you next week after I wrap up with
the Mozilla Festival. Feel free to ask more questions on list (or in repo)
and provide logs too. Once you get in a good state with this, we'd
certainly welcome pull requests to improve the documentation.

Cheers!

-- Kyle



On Wed, Oct 22, 2014 at 7:21 PM, Robert Alexander <roalexan at microsoft.com>
wrote:

>  Hello. I'm fairly new to Jupityr/Docker. Last week I did the following
> steps to get things started:
>
>
>  -Create a Ubuntu 14.04 VM on Azure # tier:basic, size:A0)
>
> -Get Docker (following official Docker instructions at:
> http://docs.docker.com/installation/ubuntulinux/)
>
> sudo apt-get update # Ensure the list of available packages is up to date
> before installing anything new
>
> sudo apt-get -y install docker.io # Install Docker by installing the
> docker-io package
>
> sudo ln -sf /usr/bin/docker.io /usr/local/bin/docker # link and fix paths
>
> # sudo sed -i '$acomplete -F _docker docker' /etc/bash_completion.d/
> docker.io - instructions say to do this, but the line is already there
>
> sudo update-rc.d docker.io defaults # optionally, configure Docker to
> start when the server boots
>
> source /etc/bash_completion.d/docker.io
>
> -Get Docker container using "Quick start" (follow instructions at:
> https://github.com/jupyter/tmpnb)
>
> sudo docker pull jupyter/demo # download Docker container.
>
> export TOKEN=$( head -c 30 /dev/urandom | xxd -p )
>
> sudo docker run --net=host -d -e CONFIGPROXY_AUTH_TOKEN=$TOKEN
> jupyter/configurable-http-proxy --default-target http://127.0.0.1:9999
>
> sudo docker run --net=host -d -e CONFIGPROXY_AUTH_TOKEN=$TOKEN -v
> /var/run/docker.sock:/docker.sock jupyter/tmpnb
>
> -Add new endpoint to virtual machine
>
> protocol:TCP, public port:8000,    private port:9999 # map internal to
> external port
>
> This wasn't without error. I got the error:
>
> APIError: 404 Client Error: Not Found ("client and server don't have same
> version (client : 1.13, server: 1.12)")
>
> So, I was attempting to re-run my steps, this time explicitly setting the
> docker version via:
>
> sudo docker run --net=host -d -e CONFIGPROXY_AUTH_TOKEN=$TOKEN -v
> /var/run/docker.sock:/docker.sock jupyter/tmpnb python orchestrate.py
> --cull-timeout=60 --docker-version="1.12"
>
> However, *now I time out trying to get to the Jupyter demo*. Did
> something change to break things between then and now?
>
>
>  -Robert
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>


-- 
Kyle Kelley (@rgbkrk <https://twitter.com/rgbkrk>; http://lambdaops.com)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ipython-dev/attachments/20141023/49cd2465/attachment.html>