Agent Communication Issues Checklist (4224297)

1. Agents not checking in at all (may show up in red)
2. Agents are checking in and inventory is not updating

Time frames of when this problem may take place are:

1. Agent inventory settings are miss configured to meet the desired number of agents in inventory.
2. The Agent count is higher than the purchased node count.
3. The appliance and agents are being upgraded to a newer version.
4. Agents did not install properly whether they were installed manually, GPO or through the provisioning tool in the appliance.
5. A configuration change was made in the security settings or Network Settings and the agents are no longer able to communicate.
6. Firewall/AV/Network device does not have the agent applications properly whitelisted.
7. Environmental factors like a Group Policy preventing communications
8. The Quest KACE Agent Service may have suddenly stopped for systems that are online and show not connected (red line through the computer icon).

Things to check when agents are having communication issues:

1. Server Version?  Agent Version(s)? Service Request number? Does the agent version work in parallel with the current server version? (example: 6.4 agent and 7.0 server version)
2. Is the agent showing as not connected (status shows computer with a line through it) or is it not completing inventory?
   IF showing a status of not connected (red line through the computer icon), is the system on?  Is the Quest KACE Agent service running?  Have you tried restarting the service using an agent command(146458)?
3. Is it a handful of agents only or is it all of them?
4. When did this issue start?
5. What steps have you already tried in resolving the issue?
6. What operating systems are involved experiencing the problem?
7. How was the agent provisioned? GPO tool / Microsoft GPO / Provisioning tool?
8. Was the server name or IP Address recently changed? If so, check to see if you need a DNS Alias to fix - (111183)
9. Verify the Firewall/AV/Network devices (such as Routers and switches) have the agent applications Whitelisted.  Disabling these and performing a forced inventory update from the agent or the appliance can rule this out as the fault.
10. Enable debug(112035) on an agent to run tests.  From the command prompt (after debug is enabled), perform a runkbot 4 0 (147398). Do you see any errors? Add them to the Service Request. IF you are already on version 7.0+ for both server and agent, this is not necessary as debug is now enabled as a new built in feature.
11. Run a Kapturestate(116144) for versions 7.2 and earlier. Kapture state (234337) for version 8.0+. Launch the Kapturestate zip file. Navigate to AgentFilesAllUsers | user | KAgent.log. (Konea log if you are on 7.0+)  Scroll to the bottom of the log and then start moving up.  What errors do you see? Please provide the Kapturestates to support with your submission if you need to create a support request. 
12. Is SSL involved? If so, are the certificates still valid?
13. Are you able to successfully Telnet to the appliance from the agent? (112029)
14. Look in the amp.conf file and see if it is blank.  If it is blank, this is a known issue for the 7.x versions and possibly 8.x.  Defect ID is K1-19581 to reference. For the most part, we have identified this issue to occur when agents are checking in at the same time a power outage takes place. This is scheduled to be remedied in a future release.