fredag 27 mars 2020

The importance of Restore

With the upcoming “World Backup Day” (31st of March), I wanted to write an blog post regarding the importance of restore since this is the most common mistake people do when working with Backup. 

And that’s the thing, we are humans and we are not flawless and therefore we make mistakes. This blog post will shed some lights over the most important part of a backup...namely restore. The other thing that is directly applicable for humans is that we are lazy and if we can rely on a piece of technology we surely will do so. The leads us to the following question, "How could you be sure that your backups are working when you need them the most?" 

The answer is quite simple, try them out!!

Throughout the years helping small, large and global enterprise companies I have noticed that there is something that all of them has in common regardless of the business size their line of business. They simply just focusing on the backups never on the restores.

Here is where I always try to make a difference. Instead of down prioritizing the Backup & Restore conversations the key point of interest should be “How can we bring back our business to an fully operative state?” The most common mistake that organizations does is not to take this matter serious enough. What it all comes down to in the bitter end is the fact that a large number of companies never get back on track with their business after a major incident or a ransomware attack since they don’t prioritize the most important parts, the recovery capabilities of their Business Services.

From the looks of it its seems very complex and hard to determine the levels of restores. I will not lie; it is not just a walk in the park, but it all comes down to what scope you associate your recovery capabilities to have. The best starting point is to start small and understand what Business Services that resides within your organisation and if applicable who is the Business Services owner. In some cases, it will be the IT department itself and in other cases it will be someone in the organization that most likely don’t have a clue what’s going on. Feels hopeless, I understand but here is where the restores tests comes into play.

For all define Business Services try to create a simple Restore Process that will restore your Business Services to a fully operational state. This is something that you of course should do in an isolated environment and never in your production. Note any deviations and mistakes you made in the first version and improve it for the next run. Do this and keep repeating until you achieve your goal of having a fully restored and fully operational Business Service. Its hard work and that is why you should document it down so you will remember it.

The technology at hand for solving this matter i very broad and you shouldn't lock your self in a mindset stating that you can't use other technologies that you already have. I encourage you to "mix-and-match" different technologies available from Microsoft to try to suite your restore demands.

Leverage the good pieces of Azure like Azure Backup Server (DPM) to protect your on premises environment and integrate it with a Recovery Services Vault (RSV) for long time archiving. Use Azure Site Recovery (ASR) for some of your Business Services if they fit the scope for ASR.

fredag 20 mars 2020

Automated backups of IaaS in Azure

Wrapped up a script that will automatically backup IaaS in Azure.
If the IaaS if tagged with NoBackup then the script will skip that IaaS since its decided it should not be included.

This script can be improved and I leave it to you to do whatever you like :)

#Authentication made via SPN
$connectionName = 'AzureRunAsConnection'

    $servicePrincipalConnection = Get-AutomationConnection -Name $connectionName
    Write-Output "Logging in to Azure..."
    Add-AzureRMAccount -ServicePrincipal `
        -TenantId $servicePrincipalConnection.TenantId `
        -ApplicationId $servicePrincipalConnection.ApplicationId `
        -CertificateThumbprint $servicePrincipalConnection.CertificateThumbprint
    if (!$servicePrincipalConnection)
        $ErrorMessage = "Connection $connectionName not found..."
        throw $ErrorMessage
        write-error -Message $_.Exception
        throw $_.Exception

#Gets all the Azure VM's
$IaaSs = Get-AzureRmVM

#Pointing out the RSV for backup and setting the vault context for editing
Get-AzureRmRecoveryServicesVault -Name "NAME-OF-SRV" | Set-AzureRmRecoveryServicesVaultContext

#Defining the backup policy
$Policy = Get-AzureRmRecoveryServicesBackupProtectionPolicy -Name "DefaultPolicy"

#Enabling backup for each IaaS object 
foreach($IaaS in $IaaSs)
     If($IaaS.Tags.Values -eq "NoBackup")
        Write-Output "Backup should not be enabled for $($IaaS.Name)"
        Write-Output "Enabling IaaS backup for VM $($IaaS.Name)"
        $RG = $IaaS.ResourceGroupName
        Enable-AzureRmRecoveryServicesBackupProtection -ResourceGroupName $RG -Name $IaaS.Name -Policy $policy

fredag 13 mars 2020

DPM 2019 failing error 812

Now and then I still bump into the 812 error when installing DPM. This error is not new but there is a common ask why this occur and how you should solve it, hence this blogpost 😊
In the very end of the installation process of DPM the installation tries to configure SQL Server Reporting Services, aka SSRS. During this step the installation fails with the error 812 stating that the mandatory encryption keys is not recreated.

System Center DPM 2019 Install Error ID: 812
Report configuration failed. Verify that SQL Server Reporting Services is installed properly and that it is running. ID: 812 . When installing Data Protection Manager DPM common installation error system admins encounter is ID: 812. This is caused by the SQL Server Reporting Services misreading the reporting database.

To solve this, logon to the Reporting Services Configuration Manager. Click the Encryption Keys followed by the Delete button. This will delete all stored connection strings, credentials and encrypted values in a subscription.

Now rerun the DPM installation and it will work as a charm!

fredag 6 mars 2020

Everything is cancelled.......:(

Due to the global circumstances of the world coping with the Corona virus all conferences has been cancelled. Its of course very sad news but it’s important to respect the decisions made by the conference owners that must apply the WHO guidelines.

sumNERV is planning for an online webinar / event where we will cover the applicable content in an online format. This will be free of charge for the community. 

Stay tuned and stay safe 😊

fredag 21 februari 2020

Experts Live Norway - Keynote

So its official!
I'm heading to Experts Live Norway 28th of May and together with my "partner in crime" Dieter Wjickmans we will deliver the Keynote for Experts Live Norway which is a great honor and also a lot of fun :)

I hope to see you there and as always lets meet up and talk, send me an email and I would love to take the moment to listen to you and help out.


onsdag 19 februari 2020

Techorama is a GO!

Just got words that I’m a selected speaker for the Techorama conference in Belgium in June.
Together with Dieter Wjikmans we will deliver a session regarding what to expect when pushing your backups to Azure.

We will ofcourse be around and avalibel for the community and as always never hesitate to reach out for a 1-1 session or talk 😊

Cu in Belgium

fredag 14 februari 2020

VHD Error 40002 not fixed in UR1

There has been a lot of voices being disappointed that Microsoft didn't include a fix for the issue "VHD Error 40002" that comes around now and then and goes away only when you reboot your DPM server.

The reason this error wasn't fixed was that its depends on so many underlying dependencies that the PG and developers need to understand all the scenario of how and why the error pops-up.

I have forwarded your concern regarding this to the product group and lets stay put and hope for a fix soon. 

torsdag 13 februari 2020

Experts Live NL

Good news just got words that I’m a selected speaker for Experts Live Netherlands!

Together with Dieter Wjickmans we will deliver a session regarding Azure backup capabilities. I will as always be available for the community during the conference and don’t hesitate to reach out if you would like to meet up for  a 1-1 talk.

onsdag 5 februari 2020

New features in UR1

With the release of UR1 for DPM comes also some new features, these are:

  • Support for ReFS volumes and ReFS volumes with Deduplication enabled
  • Windows Core Support
  • Disk exclusion for VMware VM backup
  • New cmdlet parameter

Support for ReFS volumes and ReFS volumes with deduplication enabled
With DPM 2019 UR1, you can backup the ReFS volumes and workloads deployed on the ReFS volume.

You can back up the following workloads deployed on the ReFS volumes:

  • Operating System (64 bit): Windows Server 2019, 2016, 2012 R2, 2012
  • SQL Server: SQL Server 2019, SQL Server 2017, 2016
  • Exchange: Exchange 2019, 2016
  • SharePoint: SharePoint 2019, 2016 with latest SP.

Windows Server Core support
You can install DPM 2019 UR1 on Windows Server Core 2019 and 2016.

Disk exclusion for VMware VM backup
With DPM 2019 UR1, you can exclude the specific disk from a VMware VM backup. Learn more.
Support for additional layer of authentication to delete online backup
With DPM 2019 UR1, an additional a layer of authentication is added for critical operations. You will be prompted to enter a security PIN when you perform Stop Protection with Delete data operations.

New cmdlet parameter
DPM 2019 UR1 includes a new parameter [-FragmentationPercentage]. The new parameter, calculates the fragmentation percentage for a replica, and is included in the Copy-DPMDatasourceReplica cmdlet.

tisdag 4 februari 2020

UR1 Released

It’s a great pleasure to announce that Microsoft now has released UR1 for System Center. The bugfixes included are:

  • While running the attach agent wizard when you select a computer on trusted domain and add it, the selected computer does not get listed under Selected Computers.
  • When you select short term backup to tape and change the backup mode to full, the wizard shows incorrect backup schedule option and DPM console may crash on selection.
  • In a scenario you deleted the protection group retaining the recovery point for a VMware VM, which is no more available on the ESX server. If you recreate the new VM with the same name on the same ESX instance and protect again using DPM, the recovery pane creates multiple entries for this VM.
  • The DPM Console becomes unresponsive while loading Recoverable Items for a data source.
  • DPM Service might stop during DPM maintenance jobs.
  • In some scenarios, pruning job failures leading to storage bloat
  • In addition to these, all the issues fixed in SC DPM 2016 Update Rollup 8 are also fixed in SC DPM 2019 UR1.

More information can be found here: UR1

fredag 13 december 2019

Experts Live interview

During Experts Live I got the great pleasure together with Dieter Wjickmans to be part of the Experts Live TV discussing Business Continuity and Disaster Recovery based on the Microsoft platform.

fredag 15 november 2019

Moving large quantity of data to Azure

I had an interesting project lately that I would love to share some experience from. My client had a large amount of data that needed to be moved to Azure and the bandwidth wasn’t sufficient since it would take weeks and weeks before the data was transferred. So how could we move a larger amount (more than 50 TB) of data to Azure? The answer was Azure Data Box!

Microsoft provides a great service that lets you move data from your on-premises environment to an Azure Storage account using Azure Data Box. It comes in three versions:
  • Data Box
  • Data Box Disk
  • Data Box Heavy

The technology behind the scenes are similar to the offline migration of data to an Azure RSV for DPM or Azure Backup Server.

More information regarding Azure Data Box is here

fredag 1 november 2019

UR1 postponed

A quick heads up regarding the UR1 release. As many of you already know Microsoft has released the information that they need to postpone the UR1 release for System Center 2019.

This could be of course frustrated for many of you but isn’t better that Microsoft releases a UR that works instead of just rushing to release…just saying 😉

fredag 18 oktober 2019

SQL considerations for DPM

The heart of a System Center product has always been and will always be SQL Server. This is why its so important to setup SQL Server optimal so your DPM server will deliver and optimal experience and make backups, but most importantly restore that works.

There are some key findings I always do when performing health checks on DPM (or other System Center products) and to help the DPM community I would like to share my recommendations with you.

First of is “Supported Scenarios”. This is actually a super simple thing but often missed out which leads to the consequences of you not being able to get support from Microsoft if you need it…ouch. Setting up SQL there is only one supported SQL collation, that is SQL_Latin1_General_CP1_CI_AS never deviate from this because then you deviate from Microsoft support.

Second, clustered SQL servers. Microsoft supports your DPM databases to run on remote SQL server that are clustered. However, a clustered SQL server is NOT the same as a SQL Always On setup. The latter (typing it out just in case…SQL Always On) is not supported. Now…even if its supported to run DPM on a remote SQL server this is probably the worst idea ever, don’t do it. The reason is simple, there are more unnecessary components to your DPM solution that only adds complexity…why and how is that a good idea? Guess what, it isn’t. I have had clients wanting to put the DPMDB on their production SQL hotel since this is the company standard. Don’t get me wrong, its great to standardize but you must consider the impact. A backup solution should always be “looking into” your production environment, never be part of it. Simple question, how should you restore your data when your SQL Hotel is down…guess what, you don’t.

Third, use Service Accounts (gMSA). Throughout the years people have setup their SVC accounts which is great but they all fall on the same step, changing password for all SVC accounts. “Well that is a lot of hard work so we just setup an account and make a long and complex password”, that’s the approach for to many companies. How could that possibly be a good idea? The correct setup is to leverage and use gMSA accounts that your define using PowerShell and associates with your SQL Server that hosts the DPM database.

Forth, maintenance. Always (well I have had one client that made it right) I find SQL Server not having any maintenance plans for the DPM database. Now, its important to know that from a SQL Server perspective your DPMDB is fine and consistent (DBCC check) and also your index is not to high (Re.Org Index). Those are the two most basic jobs that you must setup if you want a healthy DPMDB and a continues nice experience.

Fifth, Disk setup. Use a dedicated disk for the DPMDB that has 64K sector size. Don’t mix all the SQL Server components on the same drive as the %SystemDrive%, at least make a dedicated drive for the DB that also contains LOG, backup and TempDB.

There are more to it but this is a great starting point for you.

fredag 4 oktober 2019

TechDays meetup

During the TechDays conference I will be available to meet up the community and really looking forward to it! I got some asks regarding meetup and if you can and want to attend you are most welcome to have a community meeting during the TechDays conference on the 24th 10AM.

Just send me an email and lets meet up!!


fredag 20 september 2019

Plan a Better Disaster Recovery Plan - Adaptiva podcast

During MMS this year I had the great pleasure of being part of Adaptiva podcast where I and Dieter Wijckmans talked about our favourite subject....DISASTER RECOVERY with a little touch of Business Continuity.


Plan a Better Disaster Recovery Plan with MVPs Robert Hedblom and Dieter Wijckmans from Adaptiva Team on Vimeo.

fredag 6 september 2019

Experts Live Prague

Just got words that I will deliver both the preconference and also a session during Experts Live in Prague this year.

Together with my fellow MVP buddy Kevin Green we will try to give you all the latest and the greatest regarding System Center 2019 and also a touch of Azure. The purpose is to deliver a focused onprem pre-conf that has a lot of great content.

I have the great pleasure to deliver the session with another MVP buddy, Dieter Wijckmans, and of course will the focus be Disaster Recovery and how to set it up properly!

As always, I would love to meet up and listen to all of your thoughts in challenges and/or problems that you may face regarding System Center or Azure, don’t be a stranger 😊

fredag 23 augusti 2019

Planning for Restore not Backup

During my 20+ years of working with Business Continuity (Backup, Restore, Disaster Recovery, Monitoring etc.) and crises management I have learned a lot. The best lesson is that “The more you know the more you understand how little you can”. It maybe sounds a bit strange but understanding that you always need to develop and get better is of key essence for your professional existence.

And speaking of existence, I would love to share the most important lesson ever that I see EVERYONE getting wrong. During my years I have noticed one thing in common regarding the size of the clients I visited. It didn’t matter if they were in different fields of business nor the number of employees (size of the company). I have been, literally, all over the globe and met a lot of fantastic people and cultures. But the one thing they all had in common was they all made BACKUPS…. sweet lord. OK, lets break it down shall we. Regardless what your business does it all comes down to one simple thing…being able to RESTORE.  

Back in the days all people had one great masterplan…that was “Let’s backup everything we got one time per day since then we are safe”. So this idea is as stupid as running towards the traffic on an interstate… Being able to understand your business Recover Capabilities is the FIRST and only conversation you should have when onboarding a system for backups. How could you possibly understand the Restore Capabilities if you only provide backups? So how could you get your head around this matter?

It all starts from understanding your business and answering a simple set of questions:
  • Which Business Services do we got
  • How important are those Business Services
  • How does your Business Services relate to each other
  • How does your Business Services relate to your IT infrastructure

There are A LOT MORE to it but these questions are the most important for you to understand so you can start building your restore plans for your Business Services. There are much more to it that meets the eye, believe me…

So, how can you possibly know your Business Services, their value for the business and how they relate to each other. Wouldn’t it be great if there was a tool that you could leverage to understand the communication a server has? Well guess what, there is! Within the Azure Service Offerings you have something called Service Map that helps you understand how servers that builds up your Business Service relate to other components and applications.

Have a look at the Service Map documentation here:

fredag 9 augusti 2019

TechDays Sweden 2019

Got some great news that Microsoft wants me as a speaker at the TechDays conference in Stockholm in October. The conference will take place between the 22nd and 24st of October and more information regarding the conference is here

I will deliver a session in Swedish called “Backup, Restore och Disaster Recovery on-prem till Azure och tillbaka”. The session is about how, why and what there is to know regarding planning your restore capabilities based on both Microsoft on-premises technology and Azure.

If you attend and have questions please let me know, would love to meetup and listen to your thoughts 😊


måndag 1 juli 2019

MVP Renewal 2019

I got ”The email” in my mailbox today. Microsoft decided that I should receive my ninth MVP award in a row which I’m extremely grateful for. My ambition is always to make a difference for people that cross my road, it’s great to help people out and also provide feedback from the community to my friends at Microsoft that are truly exceptional.

I will continue to make a difference and much as I can and hopefully I we will meet you at different community meetings or conferences around the globe.

Stay tuned 😊

fredag 14 juni 2019

UR1 for System Center 2019 – Q3

Since the 2019 release the feedback from the community has been great. I know that many of you choose to send me directly emails or post your thoughts on the feedback site at MS which is great.

Microsoft will release UR1 for System Center 2019 during Q3 in 2019 which will contain and fix a bunch of problems. Its not yet official what the fixes are but as soon as Microsoft provide that information officially I will provide it to the DPM community.

fredag 17 maj 2019

MMS was a blast!!

Just got back from another great MMS. Thanks for repeatedly having me as a speaker, I’m truly honored

The best part of MMS is the small number of attendees that are present on the conference, there is a cap of less than 1000 attendees. In many cases this would be considered a bad thing but looking from the quality perspective there are much more interactions between us speakers and the attendees than other larger conferences.

I delivered four sessions and a bunch of community meet ups where the opportunity to deliver feedback to Microsoft is of key essence. The best part that I bring back home is the fact that all meetings was finished with either a “awesome, didn’t know that” or “I’m really grateful for you explaining this issue”. It great to make a difference!

Me an Dieter Wijckmans did a recording that I will share regarding my favorite topic, Business Continuity and Disaster Recovery and how you could or should leverage the available technology from Microsoft for it.

fredag 3 maj 2019

VHD Error 40002 – The VHD containing the replica or one of its snapshots could not be mounted or unmounted

One thing that was fixed in the UR7 release for DPM 2016 was the error 40002 where DPM couldn’t mount the VHD containing the replica. This was a bit of a challenge for those people who had a very tight backup schedule or backups running 24x7 (which is maybe not the right way to set it up……).

The fix back then and the fix now is to reboot your DPM Server and restart your DPM jobs. I have provided this feedback to the PG and I’m grateful for all of you providing feedback via email etc. so we can help Microsoft understand what we need from them.

fredag 5 april 2019

Your voice matters

I love the interaction between our System Center community and the Product Group at Microsoft. The best part is that they really understand us and listen to our ideas, maybe not all of them will be investigated but still they are grateful for your thoughts.

Since your voice matters its important to visit the Microsoft feedback site where you can follow the planning of what Microsoft are about to fix or is in progress fixing. Please provide your feedback here and vote for the suggestions you find the most important.

torsdag 28 mars 2019

MMS 2019

MMS 2019 is around the corner and this year I’m looking forward to deliver four sessions regarding Azure, Disaster Recovery and pushing backup strategies to the cloud.
MMS is one of my favorite conference since its rather small and the attendees are not afraid to ask questions. During this year you can (as always) meet up at any “Ask the experts” or just grab me when you see me 😊

The sessions I’m delivering this year are:
  • The Road to Azure (AKA – Ditching all of your on-prem crap)
  • Avoid the Apocalypse with a Disaster Recovery plan that works!
  • Pushing your Backup Strategy to the Cloud – What to expect and what to Gain
  • AMA – Come ask the Microsoft DC/Azure Management Team and MVP’s Anything

I have the great pleasure to co-present with Cameron Fuller (MVP), John Joyner (MVP) and Dieter Wijckmans (MVP)

See you @ MMS 2019!!

onsdag 20 mars 2019

DPM 2019 upgrade considerations

During the MVP week at Microsoft HQ in Redmond we have had discussions with the product group regarding what DPM 2019 can back up.

As many of you have already noticed on the official Microsoft webpage DPM 2019 only support Windows Server 2016 as the “oldest” version. The reason for this is due to the fact that Microsoft only provide support for the applications that resides within the mainstream support.

So before you download the ISO and start your upgrade-frenzy be very sure that you don’t miss out on any of the information regarding supported applications for backups.

fredag 15 mars 2019

SAC is no more….

As the release of System Center 2019 Microsoft also updated the information regarding the Semi-Annual Channel (SAC) is being discontinued and the Long-Term Servicing Channel (LTSC) will be around. This don’t however apply to System Center Configuration Manager (SCCM) that will have three yearly releases.

What does this actually mean? Well, to be honest it will be no difference since Microsoft provides an upgrade path from SAC versions (1801, 1807) so you will be able to upgrade your 1807 DPM server to 2019.

There are some considerations that must be covered regarding supported versions before you make any upgrade to DPM. I will try to write a blogpost explaining this a bit more thoroughly.

torsdag 14 mars 2019

Whats is new for DPM 2019

Microsoft has just relesed the information regarding the availability of DPM 2019 and the list of what the 2019 version contains has grown which is always a great pleasure!
  • Windows Server 2019 support
  • SQL 2017 support as DPM database
  • Support for newer workloads backups
  • Faster backups with Tiered storage using SSDs
  • Support for Central Monitoring
  • VMware backup to tape
  • VMWare parallel backups
More detailed information regarding all above mentioned features can be read at the official Microsoft webpage:

fredag 8 mars 2019

Experts Live Interview

Had a great experience beeing intervied regarding my sessions and also the concept regarding Backup, Restore and Disaster Recovery functions based on Azure technology.

fredag 15 februari 2019

MVP summit comming up!!

I’m heading over to Microsoft HQ in Redmond this march to attend the MVP summit but also provide YOUR feedback to the Product Group.

If you have any ideas, thoughts or information that you want me to provide to team please don’t hesitate to send me an email !

Looking forward to your emails 😊

torsdag 31 januari 2019

Breaking down IT Silos interview

Me and John Joyner was interviewed during MMS 2018 regarding our  “IT silos” session where we should proof of what needs to be done to eliminate IT silos and also showed examples how to leverage Microsoft and Azure technologies.

The interview can be viewed here:

fredag 21 december 2018

MABS V3 release notes

Since there is people asking for the location for the release notes regarding MABS V3 I have included that link below:


fredag 14 december 2018

Backup is failing after upgrade MABS v2 -> MABS v3

Recently a client upgraded their MABS (Microsoft Azure Backup Server) V2 to V3 with the result of DAG backups started to failing.

Within the release notes for the MABS V3 its stated that this could happen. To fix this issue you run the following SQL query from the SQL Server Management Studio:

IF EXISTS (SELECT * FROM dbo.sysobjects
    WHERE id = OBJECT_ID(N'[dbo].[tbl_PRM_DatasourceLastActiveServerMap]')
    AND OBJECTPROPERTY(id, N'IsUserTable') = 1)
    DROP TABLE [dbo].[tbl_PRM_DatasourceLastActiveServerMap]

    CREATE TABLE [dbo].[tbl_PRM_DatasourceLastActiveServerMap] (
        [DatasourceId]          [GUID]          NOT NULL,
        [ActiveNode]            [nvarchar](256) NULL,
        [IsGCed]                [bit]           NOT NULL
        ) ON [PRIMARY]

    ALTER TABLE [dbo].[tbl_PRM_DatasourceLastActiveServerMap] ADD
CONSTRAINT [pk__tbl_PRM_DatasourceLastActiveServerMap__DatasourceId] PRIMARY KEY NONCLUSTERED
    )  ON [PRIMARY],

CONSTRAINT [DF_tbl_PRM_DatasourceLastActiveServerMap_IsGCed] DEFAULT
    ) FOR [IsGCed]

fredag 7 december 2018

System Center Data Protection Manager 2019

Microsoft shared their roadmap during Ignite regarding what the product group was aiming for regarding the 2019 release of DPM.

This is the list that the PG is trying to accomplish:

  • Support for new versions of SQL Server and Windows Server – DPM 2019 will support new releases like Windows Server 2019 and SQL Server 2017
  • VMware to tape Backup – DPM will get support for backing up VMware VMs to tapes for long-term retention
  • Modern Backup Storage – There are a couple of improvements how DPM handles the Modern Backup Storage using ReFS. These should enhance the reliability and performance of backup storage
  • More flexibility to recovering from Azure – Even if you lose your local DPM database, you can now still restore workloads from Azure
  • Enhanced Reporting and Monitoring – Integration of DPM servers connected to Azure Backup, Azure Log Analytics or PowerBI

If you have any thoughts regarding what's included send an email and I will forward it to the Product Group :)

måndag 26 november 2018

Microsoft Azure Backup Server V3 is released!!

The following features are included in this update for Azure Backup Server v3.

Backup storage migration

By using volume-to-volume migration in MABS v3, you can move your backups on-premises for more optimized storage management. You can also move your data source to a different volume if a volume is near capacity and cannot be extended.

Prevent unexpected data loss

Assigning wrong volumes for backup storage can cause critical data loss. MABS v3 lets you disable such volumes by using Windows PowerShell. All volumes and mount points, except System Volumes, are available for MABS storage.

Custom size allocation

Modern Backup Storage (MBS) optimizes storage consumption by using storage when it is necessary. After you configure MABS for protection, it calculates the size of data that is being backed up. When multiple data are backed up together, this calculation can take a while. By using MABS v3, you can configure MABS to accept the storage volume size as default, instead of calculating the size of each file and folders that's being backed up.

Optimized CC for RCT VMs

MABS v3 optimizes network and storage consumption by transferring only the changed data during any Consistency Checks for RCT (the native change tracking in Hyper-V) VMs.

VMware VM protection improvements

VMware VM backup was in Preview mode in Windows Server 2016 that has MABS v2 installed. By using MABS v3, this is supported in production mode. MABS v3 will include support for vCenter/vSphere and ESX/ESXi licensed version 6.5, 6.0, and 5.5.

TLS 1.2 support

TLS 1.2 is a secure method of communication that is recommended by Microsoft. For MABS, TLS 1.2 is applicable for protecting workloads to cloud.

SQL Server 2017 support

MABS v3 can be installed together with Microsoft SQL Server 2017. You can either upgrade your SQL Server from SQL Server 2016 to SQL Server 2017 or use a fresh instance of SQL Server 2017 as an MABS database. By using v3, you can seamlessly back up the SQL Server 2017 workload.

Windows Server 2019 support

MABS v3 can be installed on Windows Server 2019. You can upgrade your system to Windows Server 2019 either before you upgrade or install MABs v3 or a later version.

Issues that are fixed

This update fixes the following issues:
No billing for protected collocated VMWare data sources in MABS v2.
MAB server crashes while sending billing information for a clustered data source.
Indefinite looping of RCT VM consistency check jobs because of I/O errors.
When you create a protection group, MABS displays a warning message about exceeding data limits. This issue occurs even for small data sources when you are protecting Hyper-V virtual machines.
The MABS UI stops responding after it starts if there are many recovery points for data sources.
Hyper-V backups fail because of checkpoint time-outs on the Hyper-V server.
MABS console crashes when the nightly pruning job runs for a long time.
MABS status report does not display client/laptop recovery points.
Consistency checks for Hyper-V VMs transfers more data than the size of the VMs.
When you try to recover files and folders from Azure, some items that have a path length longer than 256 characters in the online recovery point are not visible in the UI.
The UI stops responding for several minutes when a consistency check job is triggered.
Severity is lowered from “Critical” to “Informational” for the alert that is raised when an online backup fails because new disk backups are present since the last online backup.
BMR and System State data source backups consume more storage when you use Modern Backup Storage.
Alternate synchronization jobs for files and folders on a deduped volume raise failed file alerts incorrectly.  Failed files log links in failed file alerts do not show the failed files.

fredag 2 november 2018

DPM / MABS console crashes error 917

Lately I faced a challenge at one of my clients where the DPM console crashed constantly. Going through the error logs I noticed that the storage procedure was causing the error and needed to contact Microsoft support since we are not able to alter that content of the DPMDB without jeopardizing the supported scenario.

The following script was run as a SQL query from SQL Management Studio and afterwards we was able to delete protected objects without the console crashing.

Before executing anything please take note of two things:
  1. Backup your DPMDB or MABS DB before executing any SQL script
  2. You run the script on your own risk

IF EXISTS (SELECT * FROM dbo.sysobjects

    WHERE id = OBJECT_ID(N'[dbo].[tbl_PRM_DatasourceLastActiveServerMap]')
    AND OBJECTPROPERTY(id, N'IsUserTable') = 1)
    DROP TABLE [dbo].[tbl_PRM_DatasourceLastActiveServerMap]
CREATE TABLE [dbo].[tbl_PRM_DatasourceLastActiveServerMap] (
        [DatasourceId]          [GUID]          NOT NULL,
        [ActiveNode]            [nvarchar](256) NULL,
        [IsGCed]                [bit]           NOT NULL
        ) ON [PRIMARY]
ALTER TABLE [dbo].[tbl_PRM_DatasourceLastActiveServerMap] ADD
CONSTRAINT [pk__tbl_PRM_DatasourceLastActiveServerMap__DatasourceId] PRIMARY KEY NONCLUSTERED
    )  ON [PRIMARY],
CONSTRAINT [DF_tbl_PRM_DatasourceLastActiveServerMap_IsGCed] DEFAULT
    ) FOR [IsGCed]


tisdag 23 oktober 2018

Update Rollup 6

Microsoft just released the Update Rollup 6 for DPM and with that comes the following fixes:
  • Virtual Hard Disk (VHD) missing during mount or unmount operations
  • VHD mount or unmount takes a long time to finish
  • Data Protection Manager performance is decreased because of multiple update agent jobs
  • Data Protection Manager console crashes while recovering data from online recovery points
  • Data Protection Manager crashes when you move protected clients to another disk
  • Data Protection Manager System State and Bare Metal Recovery (BMR) fails intermittently for multiple servers
  • Indefinite looping occurs for consistency check jobs for Resilient Change Tracking (RCT) on a virtual machine (VM) because of I/O errors
  • Update Rollup 5 for Data Protection Manager 2016 cannot enumerate SQL Server 2017 Availability Group databases for protection
  • Data Protection Manager 2016 Update Rollup Agent causes a restart of the protected server
A fresh installation of the Update Rollup Agent could cause a restart of the protected server.

You can download Update Rollup 6 here:

fredag 12 oktober 2018

Heavy memory usage in ReFS on Windows Server 2016

For them of you who experiencing problems with ReFS I found that this article gave me some answer and if it can make someone elses life easier than I’m happy to share!

In this article Microsoft presents a way to alter the registry for Windows Server 2016 to make ReFS not becoming unresponsive.

fredag 10 augusti 2018

Experts Live in Prague

It is with great pleasure I can announce that I have been a selected speaker for the Experts Live conference in Prague this October. I will be happy to meet up with the community and hear your thoughts and deliver two sessions:

  • Real-world Restores that will Rock your world!
  • Soup to Nuts of Azure Site Recovery
 Hope to see you all in Prague 😊

tisdag 7 augusti 2018

1807 is here!

Microsoft released the version 1807 for DPM and with that comes a nice list of fixes that the productgroup has fixed:
  • Data Protection Manager status reports do not include client or laptop recovery points
  • Hyper-V backups are unsuccessful because of checkpoint time-outs on the Hyper-V server
  • Primary Data Protection Manager jobs are unsuccessful because of ongoing recovery point creation by the secondary Data Protection Manager server
  • The Data Protection Manager console crashes when the nightly pruning job runs for a long time
  • When you create or modify a protection group, you receive a warning message about exceeding data limits even for smaller data sources when you protect Hyper-V virtual machines (VMs)
  • Data Protection Manager consistency checks of Hyper-V VMs transfer data that is greater than the actual size of the VM files
  • The Data Protection Manager UI hangs after it starts if there are many recovery points for data sources
  • Data Protection Manager updates are unsuccessful when Transport Layer Security (TLS) 1.2 is enabled
Keep in mind that you ALWAYS make a backup of the DPM/MABS DB before you upgrade!

To obtain the upgrade go here