[StarCluster] Can no longer connect to IPython clusters from a remote notebook

[Justin Riley]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

FYI, IPython 0.13.1 will be included in the upcoming 12.04 AMIs.

~Justin

On 10/25/2012 05:04 PM, Alessandro Gagliardi wrote:
> Weird. I did that and I thought it wasn't working. Maybe something
> else was wrong. Regardless, it seems to be working now.
> 
> Thanks, -A
> 
> On Thu, Oct 25, 2012 at 1:15 PM, MinRK <benjaminrk at gmail.com 
> <mailto:benjaminrk at gmail.com>> wrote:
> 
> IPython 0.14 rewrites the connection/handshake step, as well as 
> serialization.  It is not compatible with 0.13, so you have to 
> upgrade all IPythons  in a given cluster (engine, controller, and 
> client) at once.
> 
> On Thu, Oct 25, 2012 at 12:50 PM, Alessandro Gagliardi 
> <alessandro at path.com <mailto:alessandro at path.com>> wrote:
> 
> I had been able to run IPython Notebook locally and connect to 
> StarCluster using: rc = 
> Client('/home/alessandro/.starcluster/ipcluster/myfourthcluster-us-west-1.json',
>
> 
sshkey='/home/alessandro/.ssh/path-ops.pem', packer='pickle')
> I upgraded IPython from 0.12 to 0.13 and it worked for the most 
> part, though I had some trouble at times, so following the advice
> at https://github.com/ipython/ipython/issues/2171 I upgraded to
> 0.14. After restarting myfourthcluster I was no longer able to
> connect. I was still running 0.13 locally and was able to generate
> the attached crash report. I noticed that my local copy of IPython
> was looking for a "url" field in the JSON file which now looks
> like: { "control": XXXXX, "task": XXXXX, "notification": XXXXX, 
> "exec_key": "XXXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX", "task_scheme":
> "leastload", "mux": XXXXX, "iopub": XXXXX, "ssh": 
> "sgeadmin at ec2-XX-XX-XX-XX.us-west-1.compute.amazonaws.com 
> <mailto:sgeadmin at ec2-XX-XX-XX-XX.us-west-1.compute.amazonaws.com>",
>  "registration": XXXXX, "interface": "tcp://XX.XX.XX.XX", "pack":
> "pickle", "unpack": "pickle", "location": "XX.XX.XX.XX" } I noticed
> that a JSON file I had for an older instance looked quite
> different: { "url": "tcp:///XX.XX.XX.XX:XXXXX", "exec_key":
> XXXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX,
> 
> "ssh": "sgeadmin at ec2-XX-XX-XX-XX.us-west-1.compute.amazonaws.com 
> <mailto:sgeadmin at ec2-XX-XX-XX-XX.us-west-1.compute.amazonaws.com>",
>  "location": "XX.XX.XX.XX" } I tried adding a "url" field but could
> not figure out what to fill it with. I then figured that the
> problem could be because of a mismatch between 0.13 on my local
> machine and 0.14 remotely, so I upgraded. Now when I try the above
> Client command, it hangs indefinitely.
> 
> Please help!
> 
> Thank you, -Alessandro
> 
> _______________________________________________ StarCluster mailing
> list StarCluster at mit.edu <mailto:StarCluster at mit.edu> 
> http://mailman.mit.edu/mailman/listinfo/starcluster
> 
> 
> 
> 
> 
> _______________________________________________ StarCluster mailing
> list StarCluster at mit.edu 
> http://mailman.mit.edu/mailman/listinfo/starcluster
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/

iEYEARECAAYFAlCKv64ACgkQ4llAkMfDcrmM6QCePaMaS1UmyMSzX9ZhJm9siudD
UHQAni+MN9O4EVhz72qzyG8jWbvJw6bW
=spm4
-----END PGP SIGNATURE-----