Agent unreachable after a while... Windows VMs in "pre login screen"

[Marius2]

Hello there,

since some weeks I have massive problem with my Horizon setup.

Here are the facts:

- Horizon 7.7 Connection Server (7.7.0.11038474)

- Separate Horizon 7.7 Composer Server (7.7.0.11038293)

- vCenter Server 6.0 (5112527)

- Windows 10 Enterprise 1703 Template with Horizon 7.7 Agent and VMware Tools 10.0.9

- Nvidia Grid / Tesla M60 vCPUs

Since I migrated from Horizon 6 to 7.7 all my VMs will end soon or later in an "Agent unreachable" state. The VMs are up and running but when I try to connect via RDP I can only see the attached screen which I named "pre login screen". This stays for about 30 seconds, then RDP session closes. After a reboot the same VM works flawless...

Any hints what this can cause?

Best regards

[sjesse]

What version of windows 10 are you using, and have you tried pcoip or blast as well?

[Marius2]

Hi there,

we are using Windows 10 Enterprise 1703.

Since now I only have used PCoIP.

[sjesse]

I'd open a ticket with support if you could, and they are going to want a support bundle from the desktop.

VMware Knowledge Base

I can't think of anything right now but something is interrupting the logon process. You could try uninstalling the horizon agent and installing it again.

If you want to look your self connect via the remote console and go to

DriveLetter:\ProgramData\VMware\VDM\logs

and look at the debug log, you should see the logon for that user and you want to see if there are any errors.

What might help too if the logon monitor is enabled and look at the logs in

C:\ProgramData\VMware\VMware Logon Monitor\Logs.

There should be an overall log and then one per logon attempt. These show the timings for the processes, you may see a spot where its stuck. The one thing that might be possible is the userinit string is broke, take a look at this as well

VMware Knowledge Base

as it shows the proper format, some times uninstalling and reinstalling the agents can break that and I think it breaks the logon process.

[Marius2]

Thanks for your reply.

In the debug log i found this entry as soon as I tried to login:

2019-04-10T14:15:40.084+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanIPC_Framework::VChanIPC_Framework(): VChanIPC CREATED 0x000001B10BCD0120, session 2

2019-04-10T14:15:40.084+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanAPI::VChanAPI(): VChanAPI CREATED 0x000001B10BC879C0, session 2

2019-04-10T14:15:40.084+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanAPI::InitServerApp(): session 2 added to map

2019-04-10T14:15:40.084+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] ipcContext::ipcContext(): ctx CREATED 0x000001B10BCCD630

2019-04-10T14:15:40.084+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanIPC_Framework::SendRegisterApp(): no wsnm channel

2019-04-10T14:15:40.174+02:00 INFO  (2044-1C34) <7220> [LogonUI] Program 'LogonUI - Windows Logon User Interface Host' started, version=10.0.15063.0 (WinBuild.160101.0800), pid=0x2044, buildtype=release, usethread=1, closeafterwrite=0, sessionId=2

2019-04-10T14:15:40.176+02:00 DEBUG (2044-1C9C) <OfflineChannel> [LogonUI] OfflineChannel::ThreadEntry(): started monitoring

2019-04-10T14:15:40.190+02:00 DEBUG (2044-1C34) <7220> [LogonUI] credfilter::UpdateRemoteCredential(): Got remote credentials, package=0, guid=60b78e88-ead8-445c-9cfd-0b87f74ea6cd, ptr=0x000001BE44201A10, size=322

2019-04-10T14:15:40.190+02:00 DEBUG (2044-1C34) <7220> [LogonUI] Sending RPC initiate shared memory request, protocol=ncalrpc, endpoint=wsnm_sharedMem, clientPid=8260

2019-04-10T14:15:55.081+02:00 WARN  (12D0-1B58) <7000> [pcoip_vchan] VChanIPC_Framework::SndMessage(): SendFastPath FAILED operation=1, sid=2, no NodeManager channel, waited=15secs

2019-04-10T14:15:55.081+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] pcoip_vchan_register_connect_cback_ex(): Register connect callback for '(null)', sid=2, hConnectCB=0x302d0

2019-04-10T14:16:10.081+02:00 WARN  (12D0-1B58) <7000> [pcoip_vchan] VChanIPC_Framework::SndMessage(): SendFastPath FAILED operation=14, sid=2, no NodeManager channel, waited=15secs

2019-04-10T14:17:21.677+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] pcoip_vchan_plugin_app_exit2(): unloaded on session 2

2019-04-10T14:17:21.677+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] ipcContext::~ipcContext(): ctx DELETED 0x000001B10BCCD630

2019-04-10T14:17:21.678+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanServerApp::~VChanServerApp(): session 2 removed from map

2019-04-10T14:17:21.678+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanAPI::Release(): session 2 removed from map

2019-04-10T14:17:21.678+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanAPI::~VChanAPI(): VChanAPI DELETED 0x000001B10BC879C0, session 2

2019-04-10T14:17:21.678+02:00 DEBUG (12D0-1B58) <7000> [pcoip_vchan] VChanIPC_Framework::~VChanIPC_Framework(): VChanIPC DELETED 0x000001B10BCD0120, session 2

The logon monitor is not enabled.

I will give it a try with uninstalling and reinstalling the agent in the master.