[Info-vax] swap and page files

[Paul Sture]

In article <kbsij9$eih$1 at online.de>,
 helbig at astro.multiCLOTHESvax.de (Phillip Helbig---undress to reply) 
 wrote:

> Back to VMS.  I have now replaced all my grey BA356 with blue ones.
> During the first change, I shut down the node while I powered off the
> box.  For the second one, I forgot.  However, apart from some expected
> mount verification, there was no problem.  How easy is it on other
> systems to power down the system disk then power it back on? 

That's an interesting question.  VMS is the only system I have come 
across in recent years which can survive a system disk going off line.

When I was running a Mac laptop booted to an external disk and no UPS 
for example, with a short power cut the laptop would survive on its 
internal battery, but of course the external disk would go offline.

If I was doing any processing at the time the laptop would lock up. If I 
wasn't doing anything I might get lucky, or not.  I concluded that if 
any system disk I/O was required (logging, paging, timestamps etc) 
during the time the disk was offline I was looking at a locked system.

With a completely cold start for both disks and system I also found the 
system could beat the disk in coming online, in which case a system 
power cycle was required.

I put your question to a Solaris/Red Hat guy yesterday and that prompted 
him to tell me the tale of a couple of clustered Solaris systems which 
have precariously long uptimes.  But that's for the other thread :-)

-- 
Paul Sture