Hello,

I have a Hyper-V 2012 Cluster using a CSV and am protecting select VMs within it using DPM 2012 Sp1.  I've applied the hotfix: http://support.microsoft.com/kb/2813630 which has resolved the major issues I was facing with backups crashing the cluster.

However, I've now noticed that every time an Express Full Backup is started I get the following error on the System log of my hosts:

Log Name:      System
Source:        VDS Basic Provider
Date:          4/11/2013 2:34:11 AM
Event ID:      1
Task Category: None
Level:         Error
Keywords:      Classic
User:          N/A
Computer:      MyHost.mycompany.com
Description:
Unexpected failure. Error code: 48F@01000003
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
  <System>
    <Provider Name="VDS Basic Provider" />
    <EventID Qualifiers="49664">1</EventID>
    <Level>2</Level>
    <Task>0</Task>
    <Keywords>0x80000000000000</Keywords>
    <TimeCreated SystemTime="2013-04-11T08:34:11.000000000Z" />
    <EventRecordID>82775</EventRecordID>
    <Channel>System</Channel>
    <Computer>MyHost.MyCompany.Com</Computer>
    <Security />
  </System>
  <EventData>
    <Data>48F@01000003</Data>
  </EventData>
</Event>

I don't seem to be having any issues with the machines pausing or crashing, but the errors are concerning to me.  When I check the state of the VSS Writers, they are all in a "Stable" state with no last error.

Any help would be appreciated.

Thanks!

--Shaun

Ditto here. Exactly the same.

Did you get a resolution?

Nope, this is still an issue.  I'm waiting for a new version of the hotfix to come out that hopefully will address this.  In the mean time I'm looking at VEEAM as an alternative to DPM, I just can't have our backup solution causing storage issues in production clusters.

Did either of you resolve this?

I am currently working through a support case with Microsoft for this exact problem, however if I am being honest I think they are unsure what is causing the error.

Lee

Nope, this is still an issue.  I'm waiting for a new version of the hotfix to come out that hopefully will address this.  In the mean time I'm looking at VEEAM as an alternative to DPM, I just can't have our backup solution causing storage issues in production clusters.

In many cases Veeam Backup & Replication and DPM can perfectly complement each other. Veeam is built for virtual environments and can extend the functionality of native solution by providing replication and other features to meet lower RTO/RPO. 

Did you ever receive an answer to this? If not, how did you resolve it.

Closing in on 2 years after your post, but I'm experiencing the exact same thing now.

Thanks, Shaun

Bill Branham branham@semcog.org or msaccount@semcog.org

Did you ever receive an answer to this? If not, how did you resolve it?

Closing in on 2 years after your post, but I'm experiencing the exact same thing now.

Thanks.

Bill Branham branham@semcog.org or msaccount@semcog.org

Did you ever receive an answer to this? If not, how did you resolve it?

Closing in on 1.5 years after your post, but I'm experiencing the exact same thing now.

Thanks.

Bill Branham branham@semcog.org or msaccount@semcog.org

Hi Bill,

I never did get an answer to this.  I've since upgraded to DPM 2012 R2 UR 4 and Hyper-V 2012 R2 and I no longer get the events in the event log.

A note of warning, upgrading the DPM agent on a client from 2012 to 2012 R2 UR3 or later WILL require the client to reboot before you can re-protect it.  

Also note, that you cannot do "online" VM backups of a guest that is on Hyper-V 2012 R2 until you update the integration services of that guest, which requires yet another reboot.  The "offline" backup WILL require that the machine be paused while the snapshot is taken.

I hope this helps,

--Shaun

HI Guy's,

i just stubeld over this thread, i have the exact samen Problem (HV Cluster with CSV on Server 2012, with DPM 2012 R2).

Somehow this seems to be "a Problem" of Server 2012.

Did Anyone of you Guy had Problems restoring a VM ? Or any real issues with the Backup?

After Playing arround a little with the Problem i found that switching CSV Owner in the Cluster Mager temporaliy fixed the problem.

But so even if MS-Support did not find a solution, investigating this seems a waste of time?

Better spend that time on a Upgrade to 2012 R2 ?

regards

Stefan

Hi Stefan,

I've been at a good working point with DPM 2012 R2 UR4 and Hyper-V 2012 R2/VMM 2012 R2 UR4 for some time now.  The time upgrading was worth it for me, even though it was a real pain due to the reboots.

Once you get to Hyper-V 2012 R2 you won't have to do side-by-side upgrades to Hyper-V anymore.  It's been announced that Hyper-V 2016 will support rolling upgrades to clusters.  So this is another compelling reason to spend time getting to 2012 R2, not to mention some of the other good features.

Now that System Center UR 6 is out I suppose its  time to upgrade again.  I talked to the DPM team and they were adamant that future update rollups for DPM would not require client reboots.  

I hope this helps,

--Shaun

HI Shaun

thanks for you reply, after reading about CU4 and that stuff i checked the DPM Agent Version.
I figured out that the DPM Agent was still at patchlevel of CU1.  The Server was at CU5.

So i updated the Server to CU6 and afterwards i updatet the Agent and did some funny live migrations to reboot the HV-Hosts ;)

Tomorrow i will check if the Error still exists. Will Post an Update than.

regards

Stefan

HI Guys

just to give everybody an Update:

Updating the DPM Angent didn't help on these Erros.

I also tried limiting paralle Backups (https://technet.microsoft.com/en-us/library/hh757922.aspx).

Still the same Erros.

regards