[Info-vax] Installing and using GNV - some feedback and questions

[Stephen Hoffman]

On 2016-10-30 18:23:55 +0000, Craig A. Berry said:

> On 10/30/16 5:58 AM, IanD wrote:
>> We might even get to a model where VMS can be totally group 
>> administered for each representative group / application (I don't think 
>> it's entirely possible now is it?, for example, how do you allow say a 
>> group manager to run jobs under a username of someone in their group 
>> only? You have to give them CMKRNL which allows them access to run 
>> under any username on the system. We need a GRPKNL privilege to limit 
>> the scope to be group bound)
> 
> You need CMKRNL to do SUBMIT/USER, but Jonathan Ridler's JUMP utility 
> allows fine-grained control over who can impersonate whom, optionally 
> logging any and all activity done under another username:
> 
> <http://vms.process.com/scripts/fileserv/fileserv.com?JUMP>

Privileges as currently implemented on OpenVMS are system-wide, and 
there's no privilege to control privilege.   That's problematic, and 
GRPKNL — or JUMP — won't help...

If implemented as I'd prefer it, privileges, logical name tables, lock 
resource domains, usernames, devices and mailboxes and the rest are 
isolated within the bundle or container and designated temporary and 
user-local storage, unless the bundle is granted broader access to the 
system.

This application isolation gets a whole lot more clear when you view 
containers as a form of OS-level virtualization, rather than the 
hardware-level virtualization many folks are using.   The latter 
approach works, but tends to be less efficient in terms of resources, 
and more costly in terms of software licenses.  OS-level virtualization 
means each bundle or container gets its own copy of OpenVMS as far as 
it can tell, but can't stomp on constructs or mechanisms or devices 
outside of of its container.


-- 
Pure Personal Opinion | HoffmanLabs LLC