[Info-vax] Life after Digital

[JF Mezei]

Today, I placed an order for an Xserve.  According to specs, it should
be just a tad faster than a Microvax II with enough memory to exceed the
32 bit limit (forcing OS-X to boot in 64 bit mode).

The power supply for my DS10L workstation failed some time ago, the
DS10L was replaced by a new Mac.

And now, the Xserve will eventually replace the "server" DS10L, as well
as the one remaining VAX running a 400-600.

Ironically, The VAX is likely to outlive the last Alpha since the
services provided by the Alpha are readily *moveable* to the Unix server
(DHCP, DNS, web, ftp, imap, pop, smtp) while the VAX still host ALL-IN-1
and provides serial ports so that functionality will need to be *ported*
(read: totally redevelopped).

It will take a while to accomplish this as I still need to sharpen my
unix skills and  really get comfortable with the booting system used by
apple (launchd).

And in the longer term, SIMH will provide my "archive" of node VELO
(vax) which started life in 1987 but got its "VELO" name in 1989.

This is the first time that I get a brank spanking new server. In the
past, I had always had used equipment. And HP got 0 dollars for me this
year where I essentially replaced my computing infrastructure.

I am unsure how to dispose of my old DEC equipment.

For the 4000-200 and later the 600 as well as a DSSI cabinet, I could
strip them down and then call up a local recycling outfit that will
gladly take that steel.

Since transportation costs would be huge for those, I doubt it would be
worth selling as a unit on ebay. Thoughts ?  Would selling individual
boards to a reseller or on ebay be worth the hassles ? It isn't so much
for the money, it is more whether it would be of help to someone needing
those parts.


The 4000-600 is a very nice machine. It isn't that much slower than the
DS10 for daily use. (but for large compiles, the difference become
visible). When I had both a 4000-200 and the 4000-600 in the cluster,
the DSSI provided some really neat functionality.

Yes, VMS clustering will be a found memory. I had been able to
transition from a microvax II in the Q5 cabinet with a 3100 workstation
to a 4 node cluster with a DS10L, 4000-200, 4000-600 (al;l 3 in the Q5
cabinet) as well as the 3100 workstation) without losing the cluster
uptime and with continued web server running ! Moving the node names
around in a running cluster was not obvious and required "reboot"
cycling to get the cluster to forget about the old node in order to
re-introduce the old node name onto new hardware.

(consider than the "in cabinet" upgrade required some work with hammers
and lots of sweat to remove a cross bar in the middle of the cabinet to
move it to the top), rejig all the clips/crews, install the solid tracks
at bottom to hold both 4000s (one at each end of cabinet) as well as
build a new "H-kit" out of fibreglass to hold the MVII's original DB25
serial ports (installed below the 4000-600).


My view of HP worsened quite a bit this year. Now, when cycling past
their offices, I see some faceleless corporation optimised to milk
customers, draw blood from employees and discard them at will. (And in
the case of their montreal (kirkland) office, they also waste much water
by directing lawn sprinklers to water the road instead of their lawn).

So, I am very glad to say that I didn't reward HP in any way for its
decisions to optimise VMS away.