[Info-vax] Where to locate software

[Chris Scheers]

Stephen Hoffman wrote:
> On 2016-06-10 02:20:51 +0000, David Froble said:
> 
>> I have run multiple applications on a single VMS system.  I've even 
>> seen multiple companies using the same VMS system.  It can be done, if 
>> the people doing it have half a clue.
> 
> That this is feasible is without question feasible.   It's getting two 
> (or more) arbitrary software packages with arbitrary dependencies to 
> arbitrarily and repeatedly and reliably install and upgrade and to 
> cleanly deinstall where this — as OpenVMS is presently implemented — 
> gets interesting.  It's certainly manually possible, but that tends to 
> delve far too deeply into the RTFM territory.  And we all know that 
> automation beats RTFM.   Further down the road from how OpenVMS 
> operates, this is also distributed updates via (for instance) RSS and 
> HTTPS and signed apps, and how vulnerable or even malicious apps are 
> isolated from each other with an effort toward avoiding wider breaches.
> 
> *This* is why I rant about PCSI and patch distribution and app isolation 
> and certificate distributions and secure password storage and package 
> management and better tools.
> 
> Because if you're doing one server, then manual processes and skilled 
> dev-ops folks can and does usually does work fine.   If you're doing 
> five servers or if you're working with products whose developers have 
> chosen to implement RTFM and (for whatever reason) not expend the effort 
> on "it just works" in their packages, this gets tedious.   If you're 
> doing fifty or five hundred servers or if you need rapid updates due to 
> security vulnerabilities or other serious issues, you're in deep 
> sneakers.   And most everything here is only going to need to happen 
> faster.
> 
> *This* is app stacking and containers and sandboxes.

Not quite entirely on topic, but:

In the 80s I set up a cluster with about a dozen boot nodes.  Each of 
the boot nodes needed to be identical.

We used a DEC product called RSM (Remote System Management?) to do this. 
  It was a pain to set up, but if you did it correctly, it automated 
distributing updates to all the boot nodes.

This is in no way comparable to how you would want to manage things 
today, but it is a problem that was thought of then and a (sorta) 
solution worked out.

Digital had it then.  <sigh>

-- 
-----------------------------------------------------------------------
Chris Scheers, Applied Synergy, Inc.

Voice: 817-237-3360            Internet: chris at applied-synergy.com
   Fax: 817-237-3074