0

I have an Azure VM Windows 10 Pro image which I use as a powerful desktop machine, accessed through remote desktop. I've been using it for about 2 months, couple of times a day (remembering to stop/deallocate it when I'm not using it, since it's 8 cores/56gb and therefore quite pricey).

On Sunday at 1:00pm I stopped/deallocated it as per usual. Then Sunday 5:00pm I restarted it, but couldn't connect.

  1. Machine was unpingable (pinging the RDP port using sysinternals/psping):

    psping mymachine.uksouth.cloudapp.azure.com:3389 
    <snip/>
    This operation returned because the timeout period expired.
    

returned timeout.

  1. In the azure portal, when I went to reset configuration (on the reset password blade for that VM), it said:

The VM agent is either unavailable, or not installed, which may prevent VMAccess from running

I tried multiple things over the next 3 hours. 1) Restarted multiple times, obvs. 2) Redeployed multiple times 3) Tried to Reset Configuration 4) Tried to uninstall and reinstall the VM access agent

Remove-AzureRmVMAccessExtension -ResourceGroupName myRg -VMName myVM -Name enablevmaccess

1 & 2 made no difference. 3 & 4 failed - hung for a while (~30 mins), then timed out. I decided to cut my losses and just set up a new Windows 10 Pro VM. I set it up, RDPed into it, then used it for an hour or so, restarted it as part of an install, and lo and behold, couldn't connect to it again, and again could see in the portal:

The VM agent is either unavailable, or not installed, which may prevent VMAccess from running

So a number of questions:

1. Am I just incredibly unlucky that 2 VMs conked out with this exact same issue or is there some generalized problem with the RDP service on the Azure portal?

2. Is what I'm doing (creating an Azure VM running Windows 10 and using it as a virtual desktop) somehow not the way Azure is intended to be used? I mean maybe your somehow "not supposed" to remote into machines all the time, just use them as servers / processors.

3. Is there any way to fix the issue on the 2 VMs I have now which are broken.

4. Is there something I'm doing that is causing the VM agent / RDP server to malfunction? (I'd be surprised if I was, I mean I used the first VM daily for over a month it worked fine. Then went duff on Sunday PM and then on Tuesday AM when I created a new one, it went duff pretty much immediately. Which feels to me like its something Azure side, not with my patterns.

EDIT

For the machine that went wrong on Sunday, I have copied some portions of the copied the serial logs here. I have copied the entry from when I logged off/stopped on Sunday when all was fine, and the immediate next entry which is from Tuesday 6:00pm. I have not cut anything out in between, its just that the log has recorded no activity between 1PM sunday and 6PM Tuesday, despite the fact that I restarted many times, redeployed maybe 3-4 times, and did all the powershell stuff.

This is from when I logged off / stopped machine successfully on Sunday afternoon.

======== Microsoft Azure VM Health Report - Start 2018-01-28T13:16:01.3045277Z ======== {"reportTime":"2018-01-28T13:15:50.5238063Z","networkAdapters":[{"name":"Ethernet","status":"Up","macAddress":"00-22-48-07-3A-91","ipProperties":[{"protocolVersion":4,"address":"10.0.1.4","isDhcpEnabled":true},{"protocolVersion":6,"address":"fe80::44fb:258a:7f04:7171%6","isDhcpEnabled":false}]},{"name":"Loopback Pseudo-Interface 1","status":"Up","macAddress":"","ipProperties":[{"protocolVersion":4,"address":"127.0.0.1","isDhcpEnabled":false},{"protocolVersion":6,"address":"::1","isDhcpEnabled":false}]},{"name":"Local Area Connection* 1","status":"Up","macAddress":"00-00-00-00-00-00-00-E0","ipProperties":[{"protocolVersion":6,"address":"2001:0:4137:9e76:14b5:3c65:f5ff:fefb","isDhcpEnabled":false}]}],"remoteAccess":{"windows":{"rdpPort":3389,"rdpEnabled":true,"rdpTcpListenerSecurityConfiguration":{"nlaUserAuthenticationRequired":true,"authenticationSecurityLayer":"TLS","protocolNegotiationAllowed":true},"rdpTcpListenerMaxConnections":3,"rdpFirewallAccess":"Allowed","rdpAllowedUsers":["MYUSER"],"rdpCertificateDetails":{"subject":"CN=MACHINENAME","thumbprint":"7A8C88CC0531A299C1CE37D57A725CA2DD9C71DB","validFrom":"2017-12-06T20:08:17Z","validTo":"2018-06-07T20:08:17Z"},"rdsLicensingStatus":null}},"accounts":{"windows":{"adminAccountPasswordExpired":false,"adminAccountDisabled":false}},"services":[{"name":"TermService","state":"Running","startMode":"Manual"},{"name":"BFE","state":"Running","startMode":"Auto"},{"name":"MpsSvc","state":"Running","startMode":"Auto"},{"name":"WindowsAzureGuestAgent","state":"Running","startMode":"Auto"},{"name":"WindowsAzureTelemetryService","state":"Running","startMode":"Auto"},{"name":"RdAgent","state":"Running","startMode":"Auto"}],"systemInfo":{"windows":{"osVersion":"Microsoft Windows NT 6.2.9200.0","isOsEval":true,"realTimeIsUniversal":true,"domainRole":"StandaloneWorkstation"}}} ======== Microsoft Azure VM Health Report - End 2018-01-28T13:16:01.3045277Z ========= I have not cut anything out here, there is just nothing in the log at all between 1PM sunday and 6PM Tuesday, despite the fact that I restarted many times, redeployed maybe 3-4 times, and did all the powershell stuff. ======== Microsoft Azure VM Health Report - Start 2018-01-30T18:05:23.8572841Z ======== {"reportTime":"2018-01-30T18:05:06.0590204Z","networkAdapters":[{"name":"Ethernet","status":"Up","macAddress":"00-22-48-07-3A-91","ipProperties":[{"protocolVersion":4,"address":"10.0.1.4","isDhcpEnabled":true},{"protocolVersion":6,"address":"fe80::44fb:258a:7f04:7171%6","isDhcpEnabled":false}]},{"name":"Loopback Pseudo-Interface 1","status":"Up","macAddress":"","ipProperties":[{"protocolVersion":4,"address":"127.0.0.1","isDhcpEnabled":false},{"protocolVersion":6,"address":"::1","isDhcpEnabled":false}]},{"name":"Local Area Connection* 1","status":"Up","macAddress":"00-00-00-00-00-00-00-E0","ipProperties":[{"protocolVersion":6,"address":"2001:0:4137:9e76:34e7:1213:f5ff:fefb","isDhcpEnabled":false}]}],"remoteAccess":{"windows":{"rdpPort":3389,"rdpEnabled":true,"rdpTcpListenerSecurityConfiguration":{"nlaUserAuthenticationRequired":true,"authenticationSecurityLayer":"TLS","protocolNegotiationAllowed":true},"rdpTcpListenerMaxConnections":2,"rdpFirewallAccess":"Allowed","rdpAllowedUsers":["MYUSER"],"rdpCertificateDetails":{"subject":"CN=MYMACHINE","thumbprint":"7A8C88CC0531A299C1CE37D57A725CA2DD9C71DB","validFrom":"2017-12-06T20:08:17Z","validTo":"2018-06-07T20:08:17Z"},"rdsLicensingStatus":null}},"accounts":{"windows":{"adminAccountPasswordExpired":false,"adminAccountDisabled":false}},"services":[{"name":"TermService","state":"Running","startMode":"Manual"},{"name":"BFE","state":"Running","startMode":"Auto"},{"name":"MpsSvc","state":"Running","startMode":"Auto"},{"name":"WindowsAzureGuestAgent","state":"Running","startMode":"Auto"},{"name":"WindowsAzureTelemetryService","state":"Running","startMode":"Auto"},{"name":"RdAgent","state":"Running","startMode":"Auto"}],"systemInfo":{"windows":{"osVersion":"Microsoft Windows NT 6.2.9200.0","isOsEval":true,"realTimeIsUniversal":true,"domainRole":"StandaloneWorkstation"}}} ======== Microsoft Azure VM Health Report - End 2018-01-30T18:05:23.8572841Z =========

EDIT 2

I just restarted the original machine. Before starting the Reset Password blade said: Now the Reset Password blade says:

The VM agent is either unavailable, or not installed, which may prevent VMAccess from running.

NO MORE ENTRIES IN THE SERIAL LOG (just like between sunday and tuesday).

Boot Diagnostics screen shot says

"Unable to get the blob content"

Also, I have attached the VM overview activity https://i.imgur.com/Yxs03fO.png - seems the machine is very busy but no network connectivity?!

I've stopped machine again now FYI.

  • What was the screenshot showing of these VMs? You can check by selecting the Boot Diagnostics tab. From there you can check to see if the VM is booting fully or not. From there you can determine how to fix it. If the VM is booting, it could be firewall or resource related. If the VM is not booting then it could be cased by windows updates that failed to install. – Micah_MSFT Jan 30 '18 at 22:43
  • You should ensure your VM is running. You should check your VM's log, like this screenshot. imgur.com/a/CEwu6 – Shui shengbao Jan 31 '18 at 2:00
  • If your VM is not running, you could not reset configuration and give error log The VM agent is either unavailable, or not installed, which may prevent VMAccess from running. The agent is running in the VM. Now, your VM is stopped. – Shui shengbao Jan 31 '18 at 2:01
  • @Shengao/@Micah. Thanks for replies. This is what I get when I click on the Boot Diagnostics / Screenshot and Boot Diagnostics / Download Serial Logs. imgur.com/a/5TmWV - frustrating! Although I can't download the whole log, I've now added some interesting parts of it to the main question. I will also now start the machines again and transcribe what happens. – ubienewbie Jan 31 '18 at 4:29
  • Do you limit port on Azure NSG, I see no network(output/input) to your VM. – Shui shengbao Feb 1 '18 at 5:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.