[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Power cycling Mark IV
I was going to answer Tom's remark about not needing a STAMP
reset about the same was as Arne. Yes if there are no bugs
in the STAMP code it will never need to be re-set. When I
design things I always assume there are and will always be
software errors. With luck the STAMP code will be under
continuos development. Many of the Mark IV systems will be
operated remotely. If there was a reset I'd build it into
the routine that handle serial communication with the STAMP.
It would send a message then if nothing came back after a while
do a reset and re-send the message. I think we could use an
unused serial control line. One of those meant for modems.
It was pointed out that the Control PC could "stop talking"
The most likely thing to happen is a failure of the real-time
control software. This can be fixed remotely.
If it is due to a hardware failure there is nothing to be done.
The OS could crash or lock up but that is highly unlikely
but if you are worried about that you can buy a hardware watchdog
card that will reboot the PC if the OS crashes. So we can
fix any software caused problem on a PC without being on site.
With a STAMP reset function we could fix any STAMP software
problem remotely also.
aah@nofs.navy.mil wrote:
>
> Tom,
> You have already stated that you haven't implemented
> all of the necessary traps to prevent the Stamp from hanging.
> Until someone writes the perfect software program, there
> will always be instances of hangs and screwups. Granted,
> most of those at this stage will be hardware problems, but
> I run the system unattended and can't be prepared to
> "take that 90 foot walk" since I don't live next to the
> telescope. Same goes for other sites, I'm sure. Why are
> you so set against implementing a remote reset for the Stamp?
> It gives the programmer a chance to either recover from
> a hardware failure, log the failure properly, or at least
> stow the system in a safe position. As I've said before,
> I don't give this particular feature high priority, but
> it seems like a reasonable feature to me.
> Arne
--
Chris Albertson home: chris@albertson-home.net
Redondo Beach, California work: calbertson@primeadvantage.com