> 	There's nothing new with sunlamp being down, it's had problems
> 	for way over a year. I find it hard to believe that it's problems
> 	are strictly software due to the knowledge of the people involved
> 	with it. I think it far more likely they were stuck with
> 	questionable hardware of indetermanite asian rim origin like most
> 	low cost x86 hardware...

Have you ever considered that PERHAPS the statement "It's having
hardware problems" wasn't always truthful or correct?


I've seen enough crashes that just can't be blamed on hardware.

For instance: networking and other code touching memory that's already
been freed.  Hardware just _Can't_ cause that, unless said hardware
just HAPPENS to DMA a pointer to a memory block on the free list into
a location where it's going to get consistently used.


there have been times in the course of the project where sun-lamp has
been stable, for many weeks at a time, under heavy load...


cgd