I’m looking at installing the IT Managed version of Omniverse Launcher on a small group of Windows hosts, and the goal is to have a small number of users be able to use these hosts to work with Omniverse and Maya.
It’s a small set - six - of computers, and I’m tasked with setting them all up with the same software, so that the people working with Omniverse can use them.
In reading the Launcher installation documentation, I find that I have a growing list of questions.
- It’s becoming clear that I should (as local administrator) install Omniverse Launcher and the requested apps and connectors, and then edit the
omniverse.toml file to define some shared areas - I’m planning to create C:\Omniverse and make library and data and cache folders there and point library_root and data_root and cache_root at those folders, respectively. Question: Is this a reasonable way to go?
- It seems to me that the .nvidia-omniverse\config folder that contains the .toml files will be located in C:\Users<name of local administrator account>.nvidia-omniverse\config - how do other users of this host access these files when they run Omniverse Launcher?
- I’m supposed to set permissions on the omniverse.toml file to be read-only so the other users can’t make changes, correct?
- One of the requested installs is the Maya 2023 Connector: will the installation process find and install the plugin files in the Maya 2023 file hierarchy, or somewhere else?
I can see that, given a different world, where Joe User has their own workstation and the ability to install things, as themselves, the configuration of Omniverse would be FAR more straightforward. That’s not the world in which we live, though, and some assistance with how to set this up so that multiple users can work (at different times) on these computers would be appreciated.
Thank you.
1 Like
Here’s an example of the multi-user issue.
As the local administrator, I installed Omniverse Launcher and crafted the omniverse.toml file to look like this:
[paths]
confirmed = true
data_root = "C:\\Omniverse\\data"
cache_root = "C:\\Omniverse\\cache"
library_root = "C:\\Omniverse\\library"
content_root = "C:\\Users\\[[username]]\\Downloads"
extension_root = "C:\\Users\\[[username]]\\Documents\\kit\\shared\\exts"
logs_root = "C:\\Users\\[[username]]\\.nvidia-omniverse\\logs"
and logged out.
Logged in as myself, and launched the Omniverse Launcher application, and it immediately pops up the Settings window, asking me to define LIBRARY PATH and DATA PATH and CONTENT PATH and CACHE PATH, and suggesting locations down in C:\Users\myusername… as if it did not read the omniverse.toml file I’d set up just a few minutes earlier.
If you have six machines, install launcher 6 times and for each one when you install launcher it will ask you for those paths. No toml file editing required. Whether it’s six machines or 60, it’s the same easy process. Do NOT edit the toml file. This would be a last resort. You can define and refine the paths in the Launcher settings.
Yes, I can specify those paths when, as the local Administrator, I install the Launcher. (The IT-Managed version, to be clear.)
But look at what I wrote: after installing the IT-Managed version of the Launcher, I logged out and logged in as me (a regular user account), and when I ran the installed Launcher, it asked me for the paths. Is this the expected behavior? Will every user on these machines need to enter those paths when they first run the Launcher application? (The documentation does not make this clear.)
Integration with Maya 2023 - already installed in C:\Program Files\Autodesk… - is requested by the users. Regular users can’t install Maya plugins - permissions - but I can, as the local Administrator. If I - as local Administrator - install the Maya Native Connector, will other users be able to see and use it?
Thanks for the reply, Richard. I do appreciate it.
I installed an app - Audio2Face - and the Maya 2023 Native Connector on this host, while logged in as the local Adminstrator - and then logged out and back in as myself (regular user account).
I can run the Audio2Face app, but Maya doesn’t load the Omniverse Connector - see the screenshot.
I’m guessing that I did not install the Connector correctly, but am at a loss as to what I should have done differently.
Because I thought it would help, I uninstalled the connector, and the older (0.20.0) version of the MayaUSD plugin and installed 0.28.0, and then reinstalled the Native Connector.
Still doesn’t show up in Maya 2023, and won’t load when I select the OmniNative.mll - says it’s unable to dynamically load that file, because “The specified module could not be found.”
AFAIK, my MAYA_MODULE_PATH isn’t non-standard.
A hopeful update:
I found that, in the local Administrator’s maya/2023 folder, there was a newly-created modules folder and a file named omni_native-208.0.0-rc.23.mod in that folder.
I logged in as me, made that modules folder - C:\Users\username\Documents\maya\2023\modules - and copied the omni_native-208.0.0-rc.23.mod file into it.
Now, when I start Maya 2023, there’s an Omniverse top-bar menu item, and an Omniverse shelf icon.
Rolling this out to users is going to be a pain - this really isn’t set up for a multi-user environment, is it?
I am confused. You are trying to install it multiple times on ONE machine or ONCE per multiple machines ? Hopefully the later.
I’m just trying to get it set up on one machine for now. Eventually there will be six machines, set up identically, and a small group of students who will be using these machines.
Richard,
Option 1 is off the table - each user has their own Windows profile.
It sounds like Option 2 is “direct the user to install the Omniverse Launcher themselves” - is that what you’re saying?
Will the users be installing their own Omniverse apps and connectors, as well?
Is there published documentation for this?
I do really appreciate your assistance with this - the person to whom I was directed as my “Enterprise support person” has ghosted me - no response in nearly two weeks.
Thank you.
1 Like
