[IPython-user] Ipython-matplotlib thread issue

Ville M. Vainio vivainio@gmail....
Thu Jul 17 09:14:26 CDT 2008


On Thu, Jul 17, 2008 at 10:41 AM, Fernando Perez <fperez.net@gmail.com> wrote:

> around line 430 in IPython/Shell.py, change the '5' in:
>
>        # first make sure the message was received, with timeout
>        received_ev.wait(5)
>
> to 10 or so.  This makes ipython wait a bit longer for the gui thread
> before declaring it dead.  It's possible that with such large
> datasets, certain operations are taking a bit long, or that you're
> going into swap.  This might help...  But I'm shooting a bit in the

Unfortunately it won't help - it shouldn't matter how long the
operations themselves will take. This timeout is only for the GUI
thread to *receive* the command. The cause probably is that the GUI
thread has hanged, or stopped calling runcode for some reason. It was
probably caused by something you called earlier - the command that you
entered immediately before the hang doesn't matter, because it never
got to gui thread and was never executed.

What you need to do is to inspect the state of the gui thread (you
have a live ipython session there, after all) and find out why it's
hanging.

The good news here is that you *do* get this message, and IPython
stays operational. In the old system, IPython would've just hung. If
you have valuable data, you can %store it and continue work in new
ipython session (until the probably bug in matplotlib gets fixed).

Perhaps you could get by with tk backend?

-- 
Ville M. Vainio - vivainio.googlepages.com
blog=360.yahoo.com/villevainio - g[mail | talk]='vivainio'


More information about the IPython-user mailing list