Managed Dark Tenant
Your isolated recovery environment in Azure and Microsoft 365, operated as a managed service by glueckkanja AG. It goes unnoticed in everyday operations, gets tested once a month, and after a ransomware attack, a single call is all it takes to bring identities, communication, and critical data back within hours rather than weeks.
Three weeks or four hours
A clean Active Directory, emergency communication for the crisis team, and access to critical data: these are the three things every recovery needs first. Building them from scratch after an attack typically takes two to three weeks, and everything else waits until they're in place. With the Dark Tenant, they're ready in advance and back up within hours and days. Rebuilding production follows in either case, just three weeks sooner or later depending on which path you took.
2 to 3 weeks: a figure drawn from our APT engagements. Phases follow the Minimum Viable Company principle.
It's 6:40 AM, and nothing works
The file server is encrypted, Teams isn't responding, and the crisis team's contact list — carefully maintained and regularly updated — happens to be sitting on that same file server. Nobody dropped the ball; everything was organized correctly for normal operations. What happens in the next few hours comes down to a different question: whether there is anywhere from which recovery can begin.
- Infiltration
- Data theft
- Encryption
- Active Directory goes down
Applications, file shares, and ERP are all waiting until Active Directory is cleanly restored.
- No communication
Email and Teams are gone, and the contact list is on the encrypted file server.
- Operations halted
Backups may also be affected, and a new Active Directory has no knowledge of any of the old permissions.
Why backups alone aren't enough
Most organizations have prepared for a worst-case scenario, and that preparation typically rests on four assumptions. Each one is reasonable, and each one holds right up until the moment an attacker takes over production. Flip the cards, and you'll see what applies from that point on.
A clean environment, ready when everything else is down
Managed Dark Tenant is an isolated recovery environment that we build and operate for you in Azure and Microsoft 365. Day to day it sits dormant, once a month we wake it up for a test run, and after a successful attack it carries the rebuild of your production environment. It consists of four modules that you can use individually or together.
Active Directory Emergency Backup & Restore
- What is protected
- Your Active Directory with all objects, permissions, and trusts that your applications, file shares, and ERP depend on.
- What is prepared
- A replicated domain controller in the Dark Tenant, Azure Backup, an APT response environment for the infected DC, and a ground-zero environment for new, hardened DCs with a fresh Entra Connect.
- In an emergency
- The most recent backup is restored in isolation and cleaned up, two new domain controllers take over, and a fresh Entra Connect reconnects to the cloud.
Azure costs included in the flat rate
Emergency Environment for Designated Survivors
- What is protected
- The ability to keep the 5 to 10 percent of your workforce operational — the people without whom nothing gets decided or communicated during a crisis.
- What is prepared
- A preconfigured Microsoft 365 tenant with no customer data, including Exchange, SharePoint, Teams, and Windows 365, populated from automated Entra ID exports of your production environment.
- In an emergency
- The Designated Survivors sign in, write, call, and work during the first hours, while production is still being repaired.
Azure costs included in the flat rate
Secured Storage for Emergency Data and Documents
- What is protected
- Emergency handbooks, contact lists, Entra ID dumps, and the data you absolutely need in the first week.
- What is prepared
- Immutable Azure Blob Storage, separate from production: once something is written there, nobody can alter it afterward — not even someone with stolen credentials.
- In an emergency
- The crisis team's contact list is stored somewhere nobody could encrypt it.
Azure consumption billed by usage
MS SQL Server Emergency Backup & Restore
- What is protected
- The SQL databases your core processes cannot restart without.
- What is prepared
- Backup dumps of the critical SQL servers in the Dark Tenant, which we restore to Azure VMs and verify every month.
- In an emergency
- The databases are restored to Azure VMs and are ready to go as soon as your applications find a clean Active Directory again.
Azure consumption billed by usage
Internally, this was once called the Azure Emergency Response Environment, or AzERE for short. The name still describes it most concisely: an Emergency Response Environment, just as a managed, dark tenant. Dark, because it sleeps until someone turns the lights on.
Two zones, and one of them stays out of reach
On the left is your production environment, on the right the Dark Tenant, which only receives from it what it needs for recovery and gives nothing back during normal operations. Use the toggle to see what changes after an attack, and click any component to find out what it does.
- AD replication over site-to-site VPN
- Entra ID export
- SQL dumps
Access to the Dark Tenant exclusively through least-privilege identities from glueckkanja and Azure Bastion.
- Entra ID
- The cloud identities of your production environment. Their state is exported automatically so that Designated Survivor accounts are ready in the Dark Tenant.
- Domain Controller
- The heart of production, on which applications, file shares, and ERP all depend. It replicates into the Dark Tenant over a site-to-site VPN and is considered untrusted after an attack.
- LoB Apps
- Your line-of-business applications. They come back once a clean Active Directory is in place again, and that is exactly what the Dark Tenant is built for.
- SQL Server
- The databases your core processes depend on. Their backup dumps land in the Dark Tenant and are test-restored to Azure VMs every month.
- Dark Tenant Core
- The dedicated, isolated tenant that underpins everything. Access is restricted exclusively to least-privilege identities from glueckkanja, with all administrative work done through Azure Bastion.
- Replicated Domain Controller
- A domain controller that runs alongside production and, in an emergency, provides the starting point for the full forest recovery. You can start and stop its VM, and that is intentionally the extent of it.
- Azure Backup
- Backups of the replicated domain controller, kept separate from anything your production environment can reach. In an emergency, we use them to restore the last clean Active Directory.
- Secured Storage
- Immutable Azure Blob Storage across three containers for Entra ID dumps, emergency documents, and data. Nothing stored there can be altered after the fact.
- Microsoft 365, preconfigured
- A preconfigured tenant with no customer data, including Exchange, SharePoint, and Teams, where Designated Survivors can communicate during the first hours.
- Windows 365
- Cloud PCs for the Designated Survivors, so nobody has to work on a device that can no longer be trusted after an attack.
- Audit and Monitoring
- We monitor sign-ins of privileged users, backup errors, VM heartbeat, the VPN connection, and the Entra ID export pipeline.
- Infrastructure as Code: Terraform from a GitHub Enterprise repository
- The Dark Tenant is built 100% as Infrastructure as Code with Terraform. Code and environment are maintained by glueckkanja, every change is traceable, and the same environment can be reproduced identically at any time.
How the emergency response works
The sequence of the five steps is defined long before anyone needs them, and that is precisely where their value lies. Scroll through or click a step to see what happens technically at each stage.
- Technical on-call hotline around the clock
- Organizational Manager on Duty
- Break-glass credentials from the customer
- No access to the compromised production environment required
- Access exclusively through Azure Bastion
- Full forest recovery
- All credentials renewed
- Forensic, AV, and hardening scans
- Persistence check
- Two new hardened Windows Servers
- Automatic promotion to domain controllers
- Third server with a fresh Entra Connect
- The disinfected DC is deleted
- Exchange, SharePoint, and Teams with no customer data
- Windows 365 as the workspace
- Accounts from the automated Entra ID exports
- 5 to 10 percent of users as Designated Survivors
- Emergency documents and data from the Secured Storage
- SQL restore to Azure VMs
- Rebuilding production on the clean AD
1. Activate
- Technical on-call hotline around the clock
- Organizational Manager on Duty
- Break-glass credentials from the customer
- No access to the compromised production environment required
First control, then communication, then data, then applications
Minimum Viable Company doesn't mean doing less. It means doing what matters first, and settling the order before the night of the attack, not during it, when everyone is exhausted and everyone has a point.
- 1Access and control A clean Active Directory, renewed credentials, and new domain controllers.
- 2Communication The Designated Survivors work with Microsoft 365, Teams, and Windows 365.
- 3Critical documents and data Emergency documents, Entra ID dumps, and SQL servers come from the Dark Tenant.
- 4Core applications Your applications run against an Active Directory that can be trusted again.
- 5Full operations Production is back, and the Dark Tenant goes back to sleep.
A restore that has never been tested is just a hope
Once a month, we wake the Dark Tenant and verify that it would deliver on its promise when it matters. We document the results, you confirm them, and that way nobody finds out for the first time on day zero whether it works.
What runs every month
- 1Domain controller restore
- 2AD Basic Integrity Test with 16 automated checks
- 3Survivor account test
- 4SQL restore test
- 5Documented results, confirmed by you
AD Basic Integrity Test
- 01DNS
- 02NTDS
- 03KDC
- 04Netlogon
- 05SYSVOL
- 06krbtgt
- 07Object counts
- 08[open] Check 8
- 09[open] Check 9
- 10[open] Check 10
- 11[open] Check 11
- 12[open] Check 12
- 13[open] Check 13
- 14[open] Check 14
- 15[open] Check 15
- 16[open] Check 16
16 of 16 passed, report sent to customer
What changes for your organization
Day to day, the Dark Tenant changes little. On your hardest day, it changes almost everything, and that is precisely the day the board, insurers, and regulators will ask about later.
- SpeedHoursActive Directory is back within hours, regardless of whether your network, server, and backup teams are reachable at the time.
- Communication5 to 10%That's the share of your staff working as Designated Survivors within the first hours, using Microsoft 365, Teams, and Windows 365.
- Proof12 per yearEvery monthly fire drill ends with a report that you confirm and that any auditor can review.
- Compliance7 out of 10NIS2 risk measures under Art. 21 that the Red Tenant and Dark Tenant support in practice, including incident handling, business continuity, and secured emergency communication. More on NIS2
What the Dark Tenant includes
Everything that needs to be in place for an emergency, we build and operate for you, and most of it is covered by the flat rate.
- An isolated environmentA dedicated tenant in Azure and Microsoft 365, built as Infrastructure as Code and fully separated from your production environment.
- A clean Active DirectoryReplicated domain controller, Azure Backup, and a ground-zero environment for new, hardened domain controllers.
- Workspaces for the crisis teamExchange, SharePoint, Teams, and Windows 365 for the Designated Survivors, preconfigured and free of customer data.
- Immutable storageSecured Storage for emergency documents, data, and Entra ID dumps that nobody can alter after the fact.
- A monthly fire drillA restore, 16 automated checks, and a report that you confirm.
- Monitoring and reviewSign-ins, backups, VM heartbeat, VPN, and exports kept under watch, plus a Designated Survivors review every six months.
- Available around the clockTechnical hotline and Manager on Duty, so a single call is all it takes to start the recovery.
- Azure costs includedAzure costs for Active Directory and the Designated Survivors environment are covered by the flat rate.
In an actual emergency, APT Response activation and time-and-materials work are added, along with Designated Survivors licenses and Azure consumption for Secured Storage and SQL.
From ramp-up to operations
- Define the rulesDefine the rulesDuring the ramp-up project, we work through design, rules, permissions, and processes with you: which services are critical, who belongs to the Designated Survivors, and who makes the call in an emergency.
- Build the environmentBuild the environmentWe set up the Dark Tenant, the site-to-site VPN, domain controller replication, Secured Storage, and the emergency environment, all as code.
- Run a rehearsalRun a rehearsalThe first fire drill shows whether everything works together as planned, before it needs to.
- Hand over to operationsHand over to operationsAfter that, Managed Operations take over, with monthly fire drills, monitoring, and a Designated Survivors review every six months, because org charts change faster than emergency plans — experience consistently bears that out.
Frequently asked questions
What CISOs, IT leadership, and admins most often want to know before the Recovery Briefing.
Is the Dark Tenant a backup?
No, your backup stays where it is and keeps its role. The Dark Tenant is the isolated environment alongside it, where a clean Active Directory, Designated Survivors communication, and your most critical data are ready and waiting so that your production restore has somewhere to begin. It is no more a replacement for a full BCDR strategy than it is for production itself.
How isolated is the environment really?
Completely, because the Dark Tenant is its own tenant. Access is restricted exclusively to least-privilege identities from glueckkanja, all administrative work goes through Azure Bastion, and you yourself can start and stop the VM running the replicated domain controller. Taking over your production environment does not give an attacker any foothold in the Dark Tenant.
How quickly are you operational again after an attack?
Within a few hours. Active Directory is back in a matter of hours, the Designated Survivors are communicating over Microsoft 365, Teams, and Windows 365 from the very first hours, and initial operational capability is reached in under four hours. How long the full production rebuild takes depends on your environment, and that is exactly what we work through together in the Recovery Briefing.
What does the emergency response cost on top?
The flat rate covers normal operations: setup, operations, monitoring, and the monthly fire drills, including Azure costs for Active Directory and the Designated Survivors environment. In an actual emergency, APT Response activation and time-and-materials work are added, along with Designated Survivors licenses and Azure consumption for Secured Storage and SQL. We say this upfront, because nobody wants to hear it afterward.
How do the Red Tenant and Dark Tenant relate to each other?
They complement each other. The Managed Red Tenant separates privileged identities and devices from production, keeping a compromised laptop from becoming a compromised domain. The Dark Tenant is built for the case where it happens anyway. One is the wall, the other is the safety net, and both work on their own as well.
Does the Dark Tenant help with NIS2?
Yes. It supports the implementation of several risk measures under Art. 21, including incident handling, business continuity, the effectiveness of risk measures, and secured emergency communication, and together with the Red Tenant that covers seven out of ten. The monthly fire drill reports provide exactly the evidence that gets asked for. More detail is on our NIS2 page.
