[Info-vax] satellite one hour ahead

[Phillip Helbig undress to reply]

In article <nv07cm$v5r$1 at dont-email.me>, David Froble
<davef at tsoft-inc.com> writes: 

> > It would be nice if fresh installations were possible without requiring 
> > a huge amount of post-upgrade configuration.
> > 
> 
> Ok, define "huge" ....
> 
> Perhaps contrary to "the grinch" I've mentioned that I find VMS rather
> easy to install. 

A vanilla install is easy.

> Just what custom stuff do you have that would be so hard to work with? 
> I'll admit, such might exist, but I'm not aware of anything very hard,
> or time consuming. 

I have three system disks in the cluster.  Stuff which is modified and 
common to all three is not on the system disks, but on a separate disk.
It's not just the logical names pointing to files.

> Now I must ask, do you have a disaster recovery plan?  Even hobbyists
> should have something like this.  If you do have such, then the steps to
> recover your desired configuration should already be defined.  If not,
> now might be the time to work up such a document? 

Sure.  First, I have HBVS.  Also, part of the reason for moving modified 
stuff off the system disk is to make it easy to clone a system disk for 
a given node from one for another node.

> I'm assuming that you have command files to set up the custom logicals
> to define your custom environment.  

Right.

> I believe, without getting into
> details, that it's been mentioned there are 19 pieces of data that would
> be better placed into a single database, with better user interface. 

Right.

> Regardless, 19 isn't such a large number.  

No.  But this is just basic VMS.  Add TCPIP and so on and there is more.

> Not saying I'm always right,
> but, if it was me, I'd have a command file set up to move the required
> common files and set up the custom logicals.  But that's just me, even
> if I'm not a full time, or even part time, system manager. 

That's the easy part.  However, there are various site-specific DCL 
procedures on the system disk which do NOT have logical names pointing 
to them.  If these are the same on all nodes, either one maintains 3 (or 
96) copies, or one moves them off the system disk, which means that each 
of these files has to be edited to execute the corresponding common file 
then exit.  A fresh upgrade to a new version of VMS might make changes 
in these files, which means reconciliation.

> Perhaps not, but I'm thinking I could do the upgrade (new install) and
> re-configure the custom environment in several hours.  And when I was
> done, it would be repeatable.  YMMV .... 

Several hours, yes.  If it works, fine.  If not, then you have a
non-working system disk.  Rollback is quick with a backup disk.  Depends 
on whether investing several hours is worth the effort.

> That written, I'll be the first to confess that some progress is called
> for. The idea of compressed libraries is one which is way past it's
> "sell by" date. And a few other things. 

Indeed.