![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
bellinghmanAny computer running multiple operating systems simultaneously has obviously to partition those systems from each other. These days it's getting pretty difficult to get code executing inside one instance of VMWare or VirtualPC or whatever to crash the host machine, even though it's partitioned in software. When you get machines built for multi partitions, doing it with hardware support, a misbehaving OS shouldn't be a problem.
So it's a little embarrassing that Sun/Fujitsu M4000-M9000 machines can apparently be crashed in such a way that not only do partitions outside the faulty one go down, but they go down so hard you need an on-site visit from an engineer before the hardware will agree to work again. At least, that's roughly what the OpenBSD guys have apparently fallen over. And moreover, it was due to attempting to make full use of the available cores - a reasonable aspiration.
(Theo de Raadt posting in bugtraq - <200809090658.m896wPlO004060@cvs.openbsd.org>)
So it's a little embarrassing that Sun/Fujitsu M4000-M9000 machines can apparently be crashed in such a way that not only do partitions outside the faulty one go down, but they go down so hard you need an on-site visit from an engineer before the hardware will agree to work again. At least, that's roughly what the OpenBSD guys have apparently fallen over. And moreover, it was due to attempting to make full use of the available cores - a reasonable aspiration.
(Theo de Raadt posting in bugtraq - <200809090658.m896wPlO004060@cvs.openbsd.org>)
