> Hmm. The SCSI spec actually demands that
> disconnecting/powring off devices must
> be possible.
>
true, but that would be at the SCSI protocol level. The low-level electrical
side of things is different.
There's an excellent article in one of the DEC technical journals
(http://www.digital.com/info/DTJ000/dtj-back-issues.htm) on parallel scsi
technology (http://www.digital.com/info/DTJR00/index.html) which goes into a
lot of detail on the electrical effects of mating/demating on the scsi bus.
quick summary:
the physical interface needs to be designed with hot-plug in mind. eg ground
must mate first, demate last. DEC's BB (breeze block) and Compaq's
proprietary drive housings are designed to be hot-swap, as are the industry
standard SCA interfaces.
the (de)mating causes a lot of disturbance on the bus, even with proper
hot-swap interfaces. in the worst case you could fry other devices on the
bus.. in the more common case the (de)mating can damage/deteriorate the
electrical interface contacts, through 'micro-arcs'..
etc.. etc.. it's a good read. (as are all the DEC tech journals).
basically: you really don't want to hot-swap normal scsi connections on a
regular basis. If you do, isolate the device from the bus before removing
the power.
-paul.
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!
