none
Backup policies configured on Windows Azure Backup are out of sync with this DPM server. (ID 33413) RRS feed

  • Question

  • Hello all,

    I keep getting this event logged in DPM 2012 SP1 as a Critical error and no matter what I do I cannot get it to disappear.

    "Backup policies configured on Windows Azure Backup are out of sync with this DPM server. (ID 33413)"

    Coupled with clickable links to "Inactivate" and "Refresh Online Policies". Neither of these seems to do anything that removes this issue from my DPM server.

    Backups works and there seems not be any problems with storage space.

    The Azure agent seems to be of latest version since there are no messages in DPM about upgrading it and no problems with Online Backup either, certificates and all are all there and works.

    Regards

    /John

    Saturday, October 19, 2013 4:34 PM

All replies

  • Hi,

    In order to investigate this, we would need a copy of your DPMDB and the logs.  Since it seems to be a benign alert, it's up to you if you want to open a support case for us to investigate.  If it turns out to be a code defect, you will not be charged for the support case. 


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Friday, November 1, 2013 2:13 PM
    Moderator
  • I just got this alert myself where Refresh online policies does nothing at all.  Were you able to resolve your issue?
    Tuesday, January 13, 2015 3:52 PM
  • Hi,

    You can manually inactivate the alert using SQL management Studio - make a backup of the DPM Database before proceeding.

    To see the alert for Azure Policy - run the first query.  If it returns any rows, run the second query.

    -- check for Azure policy alert
    select * from tbl_AHP_Alerts
    WHERE     (Resolution <> 2)
    and ErrorXml like '%1073775237%'
    
    
    -- remove Azure policy alert
    update tbl_AHP_Alerts
    set resolution = 2
    WHERE (Resolution <> 2)
    and ErrorXml like '%1073775237%'


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Tuesday, January 13, 2015 10:45 PM
    Moderator
  • I can inactivate the alert, which is what this seems to be a script, but then it comes back in 10-20 minutes.  Here are all the alerts I've gotten.  They were inactivated, then came back.

    select * from tbl_AHP_Alerts
    WHERE ErrorXml like '%1073775237%'


    EDIT: I've also tried running the "Refresh online policies" option, but it doesn't seem to do anything.
    Tuesday, January 13, 2015 11:09 PM
  • I looked through the logs Recovery Services Agent logs and saw the following warnings from today that are not in previous logs.

    0E28	5504	01/13	22:39:55.731	03	registryutils.cpp(171)		8B466D89-910B-4309-916B-EB02BF83DDF9	WARNING	Failed: Hr: = [0x80070002] : Failed: Unable to get Value HKLM\鉈߾\Software\Microsoft\Windows Azure Backup\Config\CloudBackupProvider
    0E28	5504	01/13	22:39:55.731	71	dscontext.cpp(932)	[0000000127846590]	8B466D89-910B-4309-916B-EB02BF83DDF9	WARNING	Failed: Hr: = [0x80070002] : StartUsn_\\?\Volume{e4ccbf46-96fc-11e4-bd7d-00155d01050b}_DPM_1126075830591027659: lVal : RegUtils::GetValue(REG_KEY_DLS_CLOUDBACKUPPOROVIDER_CONFIG, GetUsnRegValueName(L"StartUsn", ssVolumeName).PeekStr(), &registryStartUsn)
    0E28	5504	01/13	22:39:55.731	71	nativeutils.hpp(42)		8B466D89-910B-4309-916B-EB02BF83DDF9	WARNING	Failed: Hr: = [0x80070002] : Encountered Failure: : lVal : r.GetValue(pszKey, pT)
    0E28	5504	01/13	22:39:55.731	71	nativeutils.hpp(42)		8B466D89-910B-4309-916B-EB02BF83DDF9	WARNING	Failed: Hr: = [0x80070002] : Encountered Failure: : lVal : r.GetValue(pszKey, pT)
    0E28	5504	01/13	22:39:55.731	71	replica.cpp(642)		8B466D89-910B-4309-916B-EB02BF83DDF9	NORMAL	LC Levels: Client - 2, Service - 2
    0E28	5504	01/13	22:39:55.746	71	nativeutils.hpp(42)		8B466D89-910B-4309-916B-EB02BF83DDF9	WARNING	Failed: Hr: = [0x80070002] : Encountered Failure: : lVal : r.GetValue(pszKey, pT)
    

    Tuesday, January 13, 2015 11:31 PM
  • Hi,

    OK - if it just a nag alert and does not cause any problems, then just leave it hanging.  Else, please open a support case so we can investigate the reason for the alerts.


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Tuesday, January 13, 2015 11:44 PM
    Moderator
  • Since januari 21, 2016 we get this errors too, have never seen them before.
    And as the other says, clicking "Refresh Online Policies..." doesnt work.

    Under the Potection view all our protection groups are green and fine.

    Does anyone have any more information about this?

    Monday, January 25, 2016 7:58 AM
  • Running those queries will make the alert go away, but only for a time.  At each interval for which it checks the online configuration, it will re-raise the alert.

    I ended up rebuilding the DPM server because we had some other issues.  I have since never attempted to use the Azure Online Backup system.  IMHO, it's not reliable enough for production use.  It tends to fail to transfer data frequently, has bugs like this, and there are some workloads that it doesn't support at all (Exchange, if I remember).  Plus you need a second staging server when you restore from backup, so you have to restore from cloud, then restore again to the destination server.  It's kind of a mess and blows your RTO out of the water.

    Against all recommendation from Microsoft documentation, I stood up a DPM server in Azure as a secondary and it's working great.  DPM as a secondary in Azure is an unsupported configuration, for some reason which nobody at MS is able to explain, but it works far better than the supported solution.  YMMV.
    Friday, January 29, 2016 3:31 PM
  • Same issue here, I'm using DPM 2012 R2 CU9 and I can´t solve this error.

    Everything seems fine, but when I try to modify the protection group configuration it donesn´t work.

    Any ideas?

    best regards

    Monday, April 4, 2016 6:39 PM
  • It's running great.  It operates exactly like an on-prem secondary.  I had one issue where 1 out of 10 file synchronizations in one protection group would fail with a nondescript error and interestingly enough, MS support agreed to take the case under the condition (since it's unsupported) that if they couldn't find out why, they were not bound to any of their SLAs, which is fine by me.  The technician found that the underlying Azure storage may not be responsive in a way that DPM is designed for, but we worked around the issue by setting offsets on all the protection groups.  By reducing the amount of concurrent synchronizations with the offsets, there haven't been any further issues with synchronizations failing.

    The technician unofficially speculated that at some future point, it would become a supported configuration, it's just that there are still kinks to work out, but in my opinion, it's perfectly fine.

    Just test your restores, as always, and understand MS may not provide support.

    Since I last posted, Exchange is now a supported workload for Azure Backup, so I think it now actually supports all DPM workloads.  I still think it's not a great solution, though, for the reason that you have to "restore twice" to get data out of it.  IMHO, Microsoft is forcing things into Azure that have no business being in Azure thanks to "Cloud First" imperative from the top, leading to convoluted Azure services like this, but I digress.




    Wednesday, April 6, 2016 1:35 AM
  • I have a lot of things in Azure, so I'm not sure on the exact figure, but here's my best guess when I break it apart.

    One A3 VM: $250
    3.4TB geo storage billed (7TB configured): $273
    Virtual network for the VPN: $25
    Egress data: $3

    So that's around $551/mo.

    Based on the Azure Backup pricing calculator, it would cost me somewhere around $350/mo to do it that way.

    Other considerations for us were that we have single SQL databases that are approximately 1TB in size, which at the time exceeded the maximum size of a single instance in Azure Backup (may not be the case now) and also necessitated a gigantic staging location, which made the "restore twice" scenario less palatable.

    If they had an in-band way to restore to an agent without having to stand the DPM server back up during disaster and avoiding the necessity of the staging disk, it would be an attractive option.

    Wednesday, April 6, 2016 6:22 PM
  • I don't know if anyone has clearly answered your original question, John.

    In regards to: "Backup policies configured on Windows Azure Backup are out of sync with this DPM server. (ID 33413)"

    Cause: You have updated the MARS Agent and your Protection groups are not using the new features that came with the latest Microsoft Azure

    Solution: The solution I have found is to wait for all your backups to finish, then modify each protection group.  Press next all the way through them so they update with the new MARS features. Do this for each PG.  Once each PG is updated, the critical message will go away.

    **I am not sure why it has taken us since 2013 to come up with this simple solution.**


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread.  David Jenner IT Systems Engineer Colonial Williamsburg Foundation


    Thursday, September 22, 2016 3:05 PM