Hello,
I've noticed this hanging problem now and again, it is happening in the C
socket library. I just tried mucking around the command line with this
and managed to interrupt it (via ctrl-C) easily enough. So the problem is
how to do this via a graphical interface (the Tcl/Tk event loop is hung up
waiting for the socket so is not accepting any input, at least I think
that's what's happening). You would have thought that the socket would
time out at some point but it doesn't seem to.
Wayne
On Fri, 18 May 2007, Patrick van der Wel wrote:
> Wayne,
>
> The updateCheck script looks very nice now. By coincidence, I have another
> observation about it. Just now we had some network problems just when I
> started the script and I found out that the program was not dealing very well
> with the loss of connectivity. I even had trouble quitting it by clicking the
> window bar [X] button. Not sure this is something you want to worry about for
> the script, but I imagine it would happen the same within Analysis and could
> cause more trouble?
>
> Patrick
>
> On Thursday 17 May 2007 11:52, Wayne Boucher wrote:
> > Tim reminds me that if for some obscure reason you happen to run
> > updateCheck while Analysis is also open, then you would indeed have to
> > restart Analysis for the changes to take place there. Of course you might
> > wonder why anyone would run updateCheck with Analysis also open, but I
> > guess it's possible. If anybody tells me this is likely to happen then
> > I'll put back in another, more directed, message, into the update scripts.
> >
> > Wayne
> >
> > On Thu, 17 May 2007, Wayne Boucher wrote:
> > > Right, I've updated the update scripts so that when you run updateCheck
> > > it now says "quit" rather than "close" and indeed quits rather than
> > > closes. Also, you don't get the irritating message about having to
> > > restart for the changes to take effect.
> > >
> > > Wayne
>
|