Current K1000 Version - 6.2.109329
Current Agent Version - 6.2.1025

Anyone have advice or best practice on how to best ensure that an agent remains in good health and continues to check in with the K1000. We are a small to medium organization with approximately 800 hosts being managed in KACE over 5 different sites. Currently, MIA is configured for 30 days. A problem that occurs more often than not for us, is sometimes we'll go searching for a single host and will notice that it's been removed from Inventory. If you connect to the host, you can verify things such as the Kace Agent service is started, AMPAgent.exe is running, amp.conf file looks good.....but something at some point has caused a break in communication.

Problems compound because we are not always aware of hosts falling out of Inventory, so they end up missing patching schedules, managed installs..etc..etc.

Are there any good practices to proactively ensure maintaining good health. Right now, I'm forced to having to daily review systems that haven't checked in to the KBOX in over 20 days and then troubleshooting individually to see if users are on vacation, out of the office...etc...or if they are in the office and the host has just dropped out of Inventory. Very manual process and time consuming.

We are planning our 6.3 upgrade in the next week so I'm unsure if there's anything new in that release that will help.

Thank you all!
Answer Summary:
1 Comment   [ + ] Show Comment


  • We have created a host file entry (MAC and PC) for kbox to point to the static IP of our server, as kbox is the default address of the installer; we call this the phone home for mommy/ daddy fix.
Please log in to comment

Answer Chosen by the Author

with 6.2 and prior agents it were possible that due to different reasons the agent configuration was broken.
the config file c:\programdata\dell\kace\amp.conf is lost or corrupted if a machine crashed, or something unusual happened.
With the 6.3 agent you have 3 parts of the agent.
the amp.conf which is only used for reading out, but no more as a work file.
the amp_auto.conf which is the work file. if this is corrupted or got lost, no problem, it will be updated with check in.
and you also have the amp.conf initial config zipped.

So the easiest way is to update.
I seen also workarounds before: 
GPO which sets a working config file if the agents were affected by this effect.
(running c:\program files (x86)\dell\kace\amptools.exe resetconf host=XXX on startup)
Answered 07/17/2015 by: Nico_K
Red Belt

  • I think this is most likely going to be the best path for us. Thank you for the explanation on how the agent is changing. I was just reading about the watchdog functionality and hopefully this helps us to keep our agents functioning correctly.
Please log in to comment


We created a KACE Health script that attempts to 'auto-fix' the KACE agent.  The script is configured as a startup scheduled task.  Our organization has about 5,000 endpoints.  Over the past year it has 'auto-fixed' more than 3,000 broken agents.  Agents that it cannot fix are sent as a service request to the KACE ticketing system, and thus a technician knows to reach out and fix that host.  The script is a VB script, and the test of health is an 'Established' TCP connection on port 52230 to the KACE server after verifying that the KACE server and target port are available.  VB script was chosen because it is already present in the OS, and it is supported on Windows XP through Windows 8.x.
Answered 07/17/2015 by:
White Belt

Please log in to comment
If anyone is interested in a copy of the health script, shoot me an email.
Answered 07/21/2015 by:
White Belt

  • pls email me at
  • me as well please -
  • Hi, please send me your health script - Thanks!
Please log in to comment
Answer this question or Comment on this question for clarity