none
CMG Data Egress seems too high RRS feed

  • Question

  • We set up a CMG 4 days ago and Storage Egress is over 2 GB already.  The collection which has proper client settings deployed to it so far has only 1 PC as a member.  According to documentation, data should be about 100 - 300 MB per client per month.  How can I determine what is causing such a high amount of data egress? 

    Note: We did set up CMG with a cloud DP but no content is distributed to the cloud DP.  We only plan on using cloud DP functionality for emergency uses. 

    Thank you.

    Monday, April 29, 2019 6:31 PM

Answers

  • After deleting our CMG, we have created a new one. This time we had the guidance of a Microsoft SCCM PFE.

    Our CMG is now up and running normally and the high amount of data egress is no longer an issue.  One of the things we did differently this time was to ensure Hybrid Azure Device Writeback was enabled.  Enabling device writeback failed the first time and we had to refresh our schema.  After that everything worked normally.  I'm marking this as an answer as the problem is no longer here.

    Friday, August 2, 2019 11:45 PM

All replies

  • The 2gb would include the CMG deployment and the Client setup files that are distributed by default. 

    Is the dashboard showing anymore information? https://docs.microsoft.com/en-us/sccm/core/clients/manage/cmg/monitor-clients-cloud-management-gateway#cloud-management-dashboard

    Monday, April 29, 2019 11:16 PM
  • Thanks for the reply. I'm not sure this egress is still related to initial setup. It has gone up another GB in the last day. That makes 5 days and 3 GB of data egress....for a single client.

    Here is a screenshot of the CMG monitoring dashboard.

    Tuesday, April 30, 2019 1:14 PM
  • Still nothing conclusive from Premier Support. They are getting an Azure person to look at our data usage. It stil seems like something is wrong because the CMG service has been turned off since the first weekend and the chart still shows a solid 13GB of usage.

    Monday, June 17, 2019 7:27 PM
  • Azure billing team confirmed that over 1.5 GB of data is coming out of our CMG daily despite the fact that the service is stopped in our SCCM console. Only problem now is that they closed the Premier Support ticket again. That's twice they closed the ticket without a resolution.  If this continues, we'll have to scrap the CMG and find another way to manage devices outside our network. What a shame.
    Tuesday, June 25, 2019 1:06 PM
  • The 14 day egress adds both CMG service and storage egress. 
    Did you confirm that the CMG cloud service is stopped in the Azure portal?
    You can't stop the storage account, so its possible there are downloads coming from that

    In the Azure portal, for the CMG service and storage account service, there is a metrics section. There might be some more info there.

    Have you right clicked on the CMG in the SCCM console, then gone to properties and looked at the content tab to make sure there isn't accidentally content that has been transferred?
    Wednesday, June 26, 2019 10:50 AM
  • I'm trying to get a look at the metrics in the Azure portal but I have very little permissions and can't see the instance. 

    No content is distributed to the CMG except for the SCCM client package which was distributed by default.  We currently have no plans to distribute any content unless there is an emergency need to get something out to internet-facing clients. 

    Wednesday, June 26, 2019 3:02 PM
  • We basically hit a wall trying to troubleshoot this and elected to stop and delete the CMG from our environment. It was never utilized in production so this isn't the end of the world.

    Unfortunately we will never know what caused the high and constant data egress.

    We are planning to re-create the CMG at a point in the future with the guidance of an expert, just to make sure we didn't mess up anything the first time around.

    Wednesday, July 17, 2019 9:17 PM
  • After deleting our CMG, we have created a new one. This time we had the guidance of a Microsoft SCCM PFE.

    Our CMG is now up and running normally and the high amount of data egress is no longer an issue.  One of the things we did differently this time was to ensure Hybrid Azure Device Writeback was enabled.  Enabling device writeback failed the first time and we had to refresh our schema.  After that everything worked normally.  I'm marking this as an answer as the problem is no longer here.

    Friday, August 2, 2019 11:45 PM