SOLVED

OMS UPdate management not working at all

Brass Contributor

I got two seperate azure environments and I am trying to use oms update management (free tier).

Both environments seem to be unable to patch any server they always fail with the most helpful error message of all time "Exception from HRESULT: 0x80240007" and the completed patch management task only shows "no data" nothing unsucessful nothing successfull.OMSerror.PNG

 

Does anyone know what might cause this issue?

The documentation does not really help at all and I can't find anything about this error in azure. 

 

Regards

Carsten 

 

10 Replies

what update source are you trying to use, are you able to update the machine manually from the gui

 

I didn't have to configure any source in the updtae management solution. I am trying to update multiple windows Servers 2016 and one Windows Server 2016 which all can be updated manually from the GUI. Actually in my Visual Studio Subscription the machines actually update on every Startup. There is no WSUS Server in my deployment

so the clients are set to find updates automatically, when you log in after an update attempt does client show updates pending installation ?

 

i havent tested this the oms update, so currently i dont know much about it.

Yes  the clients are set to find updates automatically and I even had them downloaded automatically on one machine the only thing that is missing is the installation. The update job finishes but says no data and the runbook which should patch the machine only shows the error message from my original post.

I will try this feature in my test lab Thursday so I have a better understanding. could you confirm the following "Open the Windows Event Log, navigate to Application and Services Logs\Operations Manager and search for Event ID 3000 and 5002 from source Service Connector. These events indicate the computer has registered with the OMS workspace and is receiving configuration."

Thank you, I really appreciate your help. I checked one of my Windows 2016 Servers and could see both the 3000 and the 5002 events. OMS_Eventlog.PNG

 

The frequencey of the events is as low as it is because the server is in my visual studio subscription and is only active when I need it. The Servers in the other environment are already in use and I can't just log on whenever I want.

I just tried another update and checked the Job History of the Runbooks in the production environment and it seems as if the error has changed now. I didn't change the configuration but now it looks as seen below:OMS_CircularError.PNG

 

Since the runbooks are protected and I cannot see what happens inside it is really annoying.

 

So i booted up a VM in azure and connected to my OMS.

Unfortunately i did not encounter any issues.

the last screen dump you provided indicates issues downloading the update.

 

As all you VM's are based on the same MSDN image could you perhaps try a VM from the gallery ?

 

im sorry i could not provide you with a better answer.

 

have you tried the following

 

Windows update troubleshooter

 

Reset Windows update

 

Please let me know how it goes.

 

Alternativly we could try to use my OMS against one of your machines if you like.

This way we could confirm if your Client target is at fault.

Hi Kent,

 

all machines are based on the default azure Windows Server 2016 image from the azure portal.

The funny thing is that I get this error in two different subscriptions in completly differnt tenants and one of the OMS worklplaces was even setup by another colleage.

 

In the meantime I opened a Microsoft support case and we are now trying to find the cause of the issue.

 

Trying an update from your OMS is not necessary at the time but thanks for the offer.

 

I will update as soon as I know what happens here.

best response confirmed by Carsten Stahlberg (Brass Contributor)
Solution

Through a Microsoft support call we found out that the new problem is caused by a parameter error in a function call inside the oms patch script and an updated for a default printer driver from 2006 (on windows Server 2016).

 

At the moment when an update fails the script will try to add it to a failed list using this line of code:

AddUpdateToFailList GetUpdateKbID($update.KBArticleIDs) $update.Title $update.Identity.UpdateID  $null "Install Failed" $baseException.HResult 

The problem is that GetUpdateKbID and $update.KBArtcileIDs are treated as two seperate parameters.

 

By june the 15th a new version of the script should be rolled out to all azure datacenters.

The script contains a fixed function call.

AddUpdateToFailList $(GetUpdateKbID($update.KBArticleIDs)) $update.Title $update.Identity.UpdateID  $null "Install Failed" $baseException.HResult  

 

 

1 best response

Accepted Solutions
best response confirmed by Carsten Stahlberg (Brass Contributor)
Solution

Through a Microsoft support call we found out that the new problem is caused by a parameter error in a function call inside the oms patch script and an updated for a default printer driver from 2006 (on windows Server 2016).

 

At the moment when an update fails the script will try to add it to a failed list using this line of code:

AddUpdateToFailList GetUpdateKbID($update.KBArticleIDs) $update.Title $update.Identity.UpdateID  $null "Install Failed" $baseException.HResult 

The problem is that GetUpdateKbID and $update.KBArtcileIDs are treated as two seperate parameters.

 

By june the 15th a new version of the script should be rolled out to all azure datacenters.

The script contains a fixed function call.

AddUpdateToFailList $(GetUpdateKbID($update.KBArticleIDs)) $update.Title $update.Identity.UpdateID  $null "Install Failed" $baseException.HResult  

 

 

View solution in original post