Last week on PDQ Live! we went through some of the inner workings of PDQ Deploy and PDQ Inventory. If you missed it that’s ok, get ready to lift the hood and let’s get this webcast recap started.
How do PDQ Inventory heartbeats work?
How is the online/offline status determined? (1:10)
How to connect to computers with multiple IP addresses, like laptops with docks. (4:47)
Which protocols and ports does PDQ Inventory and PDQ Deploy use? (5:27)
What external connections does PDQ Inventory and PDQ Deploy make? (7:25)
Why does PDQ Inventory and PDQ Deploy access ADMIN$? (8:04)
How to troubleshoot “Target computer name mismatch”? (10:26)
Running through the deployment process. (12:57)
Why we recommend AV exclusions? (16:36)
Best practices for uninstalling an application. (17:43)
Troubleshooting “Can not connect to target share”. (22:54)
Recommendations for running packages after a Windows 10 upgrade. (25:48)
System recommendations for 20K hosts. (27:44)
Triggering computer reboots. (33:00)
Pros and cons of Central Server mode. (36:57)
We’ll now take your questions
Q: Getting “Target computer name mismatch” on a system that returned after months, no obvious DNS issues. Any ideas?
This may be due to DNS issues due to stale records.
Q: What’s the best way to uninstall an application? Is it possible to create a package?
If the application needs to be uninstalled one time, you can do so via PDQ Inventory. If it needs to be uninstalled on multiple computers, you can create a package from PDQ Inventory and deploy it to all the machines.
Q: After imaging a PC, PDQ “Can not connect to target share” for no obvious reason. Self-resolves randomly. Why could this be?
This may be due to Group Policy Objects not yet applying, thus the computers not having the correct firewall configuration.
Q: What is the best way to run a package after completing a Win 10 in-place upgrade?
Making sure that the computer has received all the updates it needs, once you see it in PDQ Inventory with the new version, you should be able to deploy to it.
Q: I’m building a new Deploy/Inventory central server for 20K hosts. What are your CPU/RAM recommendations?
With those many hosts, we recommend having the database on fast drives, preferably SSD as there will be a lot of reads and writes. This video goes through some configuration changes for large environments.
Q: Is there a way to trigger a reboot on a computer-based on the scan status?
There’s not a native way to trigger based on scan status, but we do have a collection “Reboot Required”, which lists computers that need reboots based on scanner results. You can create a package with a reboot step to target this collection.
Q: What are the pros and cons of using Central Server with only a few users?
If multiple administrators will be managing all the computers, then Central Server is the way to go. If each administrator is in charge of their own group of computers, then keeping it separate may be better.