Some users may be experiencing issues when trying to access customer resources like the Case Portal or the Product Licensing Center. Our team is aware of the issue and is working to resolve it. Click here for more information.

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Flexera inventory agent - not picking up Applications - Mac OS

we're deploying Flexera inventory agent on both Windows / Mac OS. Windows OS was pretty straightforward, didn't had much issue as devices/apps get picked up after agent install.

headache comes to Mac: finally got a way to deploy mgsft_rollout_response to the /var/tmp folder. From the Flexera portal, MacBooks gets picked up, BUT it's an empty shell. Flexera only picks up the device with no Applications installed(only shows Mac OS as an App). Why is it only picking up Mac OS as the installed app? If the device shows up on Flexera I'd assume it'd work, not picking up an empty shell.

I then tried to run a manual inventory scan: which surprisingly works, the MacBook that runs the below command gets all Apps uploaded to the Flexera portal. Macs that didn't run this, is showing an empty shell. Is the inventory scan turned off by default, any controls of this?Any idea?

cd /opt/managesoft/bin
sudo ./ndtrack –t machine –o upload=true

(1) Solution

The deployment of the Windows agent allows you to use the bootstrap file to initiate a full inventory scan upon agent installation.

Mac (and Linux for that matter) does not have that capability.  If you leave things as they are, you'll see what I call a "hardware check in" (what you call "an empty shell") and the log of that scan being written to mgs1-tracker.log. (Full scans get written to tracker.log.)

You could either

  • Wait for the scheduled software scan to run based on whatever schedule you set up, and then wait for Reconciliation to process the inventory
  • Script a run of ndtrack that performs the full inventory shortly after the agent installation completes

Our Linux installs are scripted.  The script verifies that the install completed and then kicks off a background job that waits an additional 30 minutes before launching the full inventory.

--Mark

View solution in original post

(5) Replies

The deployment of the Windows agent allows you to use the bootstrap file to initiate a full inventory scan upon agent installation.

Mac (and Linux for that matter) does not have that capability.  If you leave things as they are, you'll see what I call a "hardware check in" (what you call "an empty shell") and the log of that scan being written to mgs1-tracker.log. (Full scans get written to tracker.log.)

You could either

  • Wait for the scheduled software scan to run based on whatever schedule you set up, and then wait for Reconciliation to process the inventory
  • Script a run of ndtrack that performs the full inventory shortly after the agent installation completes

Our Linux installs are scripted.  The script verifies that the install completed and then kicks off a background job that waits an additional 30 minutes before launching the full inventory.

--Mark

That explains why. Thanks. From where do I set the schedule scan to run? I don't see it anywhere in the bootstrap file

You can't set the schedule to run in the bootstrap file.

You set it on the FNMS server under Discovery & Inventory > Settings, scroll down to Inventory Agent Schedule.

tempo.png

The Inventory Beacons will receive this change and when the agent asks for any policy updates, the schedule change will be saved and then followed by the agent.

--Mark

Is there any way of checking the current Schedule settings on the macbook itself?
Something like, the schedule settings saved in an .ini file, that i can retrieve and double check what's the current schedule settings?

You could examine the tracker.log file and see when the agent kicked off its full inventory collection.  Each run of the agent appends to this file so you could simply examine the first line of each run to determine the approximate inventory collection schedule.

You could also look at schedule.log which will show you when certain components of the agent were run (schedule updates, policy updates, generate inventory, etc).  I have the 2023R2 version of the agent and it writes to this log file; not sure in what version this type of logging started.

I suppose you could also read the raw .nds file that gets downloaded by the agent and forms the basis for the schedule.

 

--Mark