[Xastir] Mac: odd process problem

[Chip G.]

On Jul 4, 2007, at 09:02, Jason Winningham wrote:

> On Jul 4, 2007, at 4:30 AM, Chip G. wrote:
>
>> So far I've changed two things. I recompiled with a stable  
>> release. I installed an updated USB driver. We'll see if the two  
>> help. If everything is stable I'll try going back to the latest  
>> release. If things are still stable then it would seem to indicate  
>> the USB driver.
>
>
> One other thing:  there are generally two device special files for  
> each serial port - /dev/tty.whatever and /dev/cu.whatever.  Various  
> xastir-on-macOS users have had trouble with /dev/tty.* with some  
> devices and drivers.  Using the keyspan, I've never had any trouble  
> with either.  You might try /dev/tty. instead of /dev/cu. and see  
> if that makes a difference.
>
> Oh, and I had a really fun time, crashing the kernel and  
> everything, when I mistakenly used /dev/ttty.* _and_ /dev/cu.* at  
> the same time.  It wasn't a keyspan, and that driver caused the  
> kernel to dump core.  It's not inconceivable that some other  
> application has opened /dev/tty.* and you are using the /dev/cu.*  
> hook into the same device, hence strange problems.

More good words. I was one of those people for whom the tty.* drivers  
never seemed to work properly. That was one of my first early  
struggles with Xastir. I have no application, that I know of, which  
uses the tty.* Keyspan driver. I only have one other ham Unix program  
and I know it uses the cu driver and will complain if I try to start  
it at the same time.

It made it at least 4.5 hours, but it did eventually freeze. So it  
wasn't the bleeding edge release since the stable release is having  
the same problem. The latest version of the USB driver has the same  
problem. I guess the next step will be to restart it with that turned  
off for while and see if it can go 24hrs without crashing.


73,
--de Chip (N1MIE) FN41bn