[R] R kicking us out

[(Ted Harding)]

On 13-Jun-06 Erin Hodgess wrote:
> Dear R People:
> 
> I am using R for teaching purposes in a large classroom.
> 
> Each computer has its own copy of R.
> 
> However, every once in a while, about half of us will get 
> thrown out of R, for no apparent reason.
> 
> By the way, it has happened in other classrooms as well.
> 
> 
> Has anyone else run into this, please?
> 
> If so, how have you solved this problem, please?
> 
> Thanks in advance!
> 
> R for Windows

Hmmm ... you're asking statisticans here!

When it happens, is it at about the same moment for all
the machines it happens on?

Is the set of machines it happens on different every time,
or do some seem to be systematically vulnerable?

(As you can tell, I've not experienced this, but am groping
for a mechanism. I find "once in a while, about half of us
will get thrown out of R" puzzling as a phenomenon -- strongly
suggestive of non-independence of machines, despite each
machine having its own copy of R).

Presumably they're networked. Might it be possible that (e.g.)
a central server is re-profiling some of the machines from
time to time (or similar)?

If it's a group phenomenon, I'd be tempted to look either in
the hardware direction (e.g. power fluctuation to which some
machines are sensitive and get e.g. corrupt bytes in RAM
leading to segfault or the like), or the software direction
(e.g. interference from external server as above, or maybe
one machine making a request to others over the net).

Though each machine "has its own copy of R", does each one
have its own copy of the full set of packages that you use?
Or is there some sort of central repository for packages
from which machines can pull down extras as they need them?

Is there a central "monitoring" machine in the classrom for
the instructor to use to collaborate with [groups of] students?

Just random thoughts, really! But hoping that they may help.

Best wishes,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 13-Jun-06                                       Time: 21:29:59
------------------------------ XFMail ------------------------------