[Info-vax] Databases versus RMS

[John Wallace]

On Apr 20, 10:11 pm, Jan-Erik Soderholm <jan-erik.soderh... at telia.com>
wrote:
> Phillip Helbig---undress to reply wrote 2012-04-20 21:35:
>
>
>
> > Right.  This is why, in addition to HALT, there are both BOOT and
> > RESTART options at the console.  These control what happens after a
> > power cycle.  HALT stays at the console prompt, BOOT boots and RESTART
> > tries to pick up where it left off, with the assumption that the
> > contents of memory are still good...
>
> I guess that RESTART is a left-over from the "core" days.
> Core memory keep it's data even without power. :-)
>
> Jan-Erik.

Restart in modern context is nothing to do with core (or even battery
backed) memory. RESTART actually means "see if there's a crash dump in
memory, and if there is, write the dump before rebooting". I meant to
post this in context of a discussion a week or two ago but probably
didn't. Definitive texts include:
http://h71000.www7.hp.com/doc/83final/ba322_90045/apas02.html which
says
"RESTART ... forces the system to attempt to write a crash dump to the
dump file, and after the dump write completes, the system tries to
reboot itself automatically."

Also, thanks to Philip for actually mentioning a part number. Prior to
the Storageworks implementation, there was also the ESE20 and ESE50,
but I couldn't find any easily locatable info on these; I guess they
largely pre-date the Internet.