> ----- Original Message -----
> From: David T. Lewis
> Sent: 02/19/14 07:54 PM
> To: The general-purpose Squeak developers list
> Subject: Re: [squeak-dev] OsProcess - UnixProcess Hangs - VPN related
> 
> On Wed, Feb 19, 2014 at 08:29:20PM -0500, JohnReed Maffeo wrote:
> > I am currently working in an environment where I have to access external resources using VPN.
> > 
> > I was working in Squeak (4.4-12327) last night, worked with VPN enabled today, closed it launched Squeak and now when I execute "osp := ExternalUnixOSProcess command: cmdLine" and instpect osp, it is undefined.
> > 
> > When I execute "UnixProcess testRunCommand" I get the warning "process accessor module not available"
> > 
> 
> This is the error message that you will see if you run your image on a VM
> that does not have the OSProcess plugin. You can confirm this by evaluating:
> 
>  OSProcess accessor canAccessSystem ==> true

The image that I am running in has OSProcess installed and working for several months. The code that fails only fails in conjunction with my use of Cisco VPN software. 
> 
> 
> > I don't have a complete test case defined of the exact repeatable steps but, I think the issue resolves itself
> > when I reboot.
> > 
> > Is there a way for me to get UnixProcess to wake up? I tried to initialize ??it, but that did not help.
> >
> 
> Perhaps you are running your image on a different VM, or on a different operating
> system? If you run it on a VM with the OSProcess plugin, I would expect it to return
> to normal.

> 
> But I do not understand the part about the issue resolving itself when you reboot
> the system. So I think maybe I am misunderstanding the problem.

If I reboot my computer, launch Squeak and run my program, OSProcess works as expected. The problem is associated with launching VPN while I have an image open. I have been spending my development time trying to get my application working well enough so that I can spend some time digging into the root cause of this particular issue. That happened last night and I spent some time poking around and discovered that "HTTPSocket initialize" seems to resolve the issue, if so, this is not an OSProcess issue.

I will have to do more testing to verify the specific failure and recovery scenario, but I think I am on the right track. Now that I have a working application, I can focus on Socket and see if I can figure out what is happening.

> 
> Dave

jrm