One million pageviews in first 21 months

Back to ConfigMgr main menu

I started this blog in February 2013 and I've really enjoyed sharing tips and ideas with the community. I've also received overwhelming support and I really appreciate it. I'm amazed to see that there have been over one million pageviews in the first 21 months.

I will continue to post blogs about subjects that interest me. I hope they will be of interest to others. My focus for 2015 will be on Microsoft EMS and Azure.

Thanks again,
Gerry



 

Intermittent issue with Direct Access connection

I had this tricky problem on a customer site recently. Direct Access (Windows Server 2012 R2) worked perfectly most of the time but intermittently the Workplace Connection would just remain "Connecting" indefinitely. The logs showed that the connection failed with the following error:

0x274c

CouldNotConnectToDirectAccessServer

This error was too generic to be useful and there was no pattern to this bizarre behaviour.

I eventually found the problem. ISATAP had not been removed from the Global Query Blocklist on one of the DNS servers.

Note that this is not an automatic process. It has to be done manually on all DNS servers. I removed ISATAP and restarted the DNS Server service.

Problem solved.

Linux Management with ConfigMgr - Issues I've encountered

Back to ConfigMgr main menu

Back to Linux menu

I encountered a couple of issues while carrying out this work. I hope that they can save you some troubleshooting time.

1. Extracting client installation files.

The "ConfigMgr Clients for Linux.exe" file is 62MB.

This file extracts into the client installation files that you require. 

 You must extract this file using an x64 OS.

You will receive this error if you try to extract using x86 OS.


2. Software Distribution.

I encountered error 10006 while deploying an RPM package to a Linux Server (I was trying to deploy Teamviewer for Linux). After quite a lot of troubleshooting I finally tried to manually install the software on the server and found that the installation failed due to missing prerequisite files. Clearly my ConfigMgr deployment would never have worked and it's very difficult to determine the cause from log files. 

Linux client - software distribution and patching

Back to ConfigMgr main menu

Back to Linux menu 

Previously we installed the ConfigMgr client and the Linux server is now managed and reporting normally. We are now ready to deploy software to the server. No further configuration is necessary. We should be aware of the following:

1. The ConfigMgr client will use the Network Access account to connect to the Distribution Point and download the package to the ConfigMgr cache on the Linux server.
2. You must configure the following option for your deployment - "Download content from Distribution Point and run locally".
3. You must configure the deployment as "Required". "Available" is not supported for Linux servers.
4. You do not need to modify the permissions of an RPM file to allow it to be executed.

RPM (RPM Package Manager) is a popular utility for installing software on Unix-like systems, particularly Red Hat Linux. Download your RPM package to a source folder.

(Tip - carry out a manual test installation of the RPM package before you deploy it in production. If the RPM has some missing prerequisites the deployment will fail with error 10006. It's hard to troubleshoot this but easy to identify in a test.)

We use the traditional package/program for this patching/software distribution.

Right click Packages and choose to "Create Package".

Name the package and configure the location of the source files.

Choose "Standard Program".

The syntax of an RPM installation command is as follows:

rpm -i filename.rpm

Click Next to finish.

Don't forget to distribute the content and then you can deploy the package.

Choose a collection that contains the Linux server.

You MUST choose "Required".

You MUST choose "Download content from Distribution Point and run locally".

Carry out a quick test on the Linux server to verify that the patch is not already installed.

rpm -qa | grep ppl-0.10.2-11.el6.x86_64 

(note that the file extension .rpm is not included in this command).

Request Machine Policy 

 /opt/microsoft/configmgr/bin/ccmexec -rs policy

 Test again

rpm -qa | grep ppl-0.10.2-11.el6.x86_64 

Patch is now present.
 

Successful deployment.

Linux client - machine policy and hardware inventory

Back to ConfigMgr main menu
 
Back to Linux menu

Previously we copied the ConfigMgr client installation files to a Linux server and installed the client. However, the default setting for a ConfigMgr site is to automatically approve only computers in trusted domains.

Our Linux server clearly is not in a trusted domain so we must manually approve it.

Search for the Linux server in All Systems. Right click and choose Approve.

Select Yes to accept.

Linux server is now approved. At this point the Linux server is now managed by ConfigMgr. The server will get policy according to the schedule you have configured. We can force the download of the policy (in the same way that we can initiate "Machine Policy Retrieval" on a Windows device).

Request Machine Policy by executing the following command:

/opt/microsoft/configmgr/bin/ccmexec -rs policy

Perform hardware inventory by executing the following command:

/opt/microsoft/configmgr/bin/ccmexec -rs hinv

Examine the end of the ConfigMgr log file:

tail -F /var/opt/microsoft/scxcm.log

(Tip: press ESC & q together to leave the log view and return to command line).

See the successful inventory in the MP_HINV.log file.

Inventory of the Linux Server.

In the next section we will deploy a Linux patch to the server via ConfigMgr.