Digital workplace suite for EU data sovereignty

A self-hosted, EU-operated workplace suite removes the US-jurisdiction exposure of productivity clouds only if every component, support path and admin channel is also kept in scope. This page sets out what openDesk or a Nextcloud-centred stack covers, where the gaps are, and how to run it sovereign.

On this page

Why a digital workplace suite for EU data sovereignty needs more than a hosting change

A self-hosted EU suite such as openDesk or a Nextcloud-centred stack can meet the goal of reducing dependence on US-controlled productivity clouds, but only if hosting, vendor support access, identity and software updates all stay inside the EU operating boundary. Moving the servers to an EU data center while the administrators, support engineers or update channels sit outside the EU leaves the organization exposed. The Schrems II judgment and EDPB Guidelines 05/2021 make this concrete: remote access from a third country, even only displaying data on a screen for support or administration, counts as a transfer under GDPR Chapter V, and so does storage in a cloud outside the EEA. A support contract that grants a non-EU engineer screen access to ePHI, personnel files or case data is therefore a transfer decision, not just a procurement detail.

The short answer on platforms is that openDesk covers the broadest set of workplace functions (files, office, groupware, chat, video, wiki, projects and identity) in one Kubernetes-based stack, while a Nextcloud-centred stack is smaller and easier to scope if you only need files, office and collaboration. Both have gaps to plan for, and each bundled component carries its own license and support path. The decision your team should make first is an operating boundary: who can reach the platform, from where, under which support contract, and which update sources you trust. Confirm these with each vendor and integrator in writing before you choose the stack, because the answer changes the design more than the choice of server location does.

Which transfer, access and portability rules shape the build

This table maps the transfer, remote access, government access and switching rules to concrete build requirements for a self-hosted suite. Read each row as a rule, the build decision it forces, and the evidence an assessor or auditor will ask you to produce. The main takeaway is that sovereignty depends on how the platform is operated, not only on where it is hosted.

RuleBuild requirementEvidence to keep
GDPR Article 44 and Chapter V (transfers, including onward transfers)Map every component, support path and backup target to a location. Block any path that places personal data in a third country without a valid transfer basis.Data flow register listing each component, hosting location, operator and any onward recipient
EDPB Guidelines 05/2021 (remote access counts as a transfer)No third-country remote support or screen sharing into production. Restrict vendor and support access to named EU staff, with access logged and reviewed.Support access policy, session logs and a list of permitted access locations
EDPB Guidelines 05/2021 (Example 12, EU processor under third-country law)Assess whether any EU-based provider or subprocessor is subject to third-country law with extraterritorial reach before engaging it, and document the exposure.Written assessment of each provider's legal exposure and the conditions for compliance with foreign requests
GDPR Article 48 (third-country disclosure orders)Contract for a clear process where any foreign order is reviewed against a mutual legal assistance treaty or equivalent agreement before data is disclosed.Contract clauses, internal escalation procedure and a record of requests received
EDPB Recommendations 01/2020 (supplementary measures)Where a provider could access data in the clear under foreign powers that exceed what is necessary, plan for encryption with keys held in the EU or a different hosting model. Confirm the approach with counsel.Key management design, key location record and transfer impact assessment
CLOUD Act, 18 U.S.C. 2713 (US provider possession, custody or control)If any component, support contract or backup service is controlled by a US provider, treat its data as reachable regardless of server location. Prefer operators with no US parent or US-law exposure for the stack.Vendor ownership and control review covering parent companies and support staff nationality or location
Data Act Article 32 (non-personal data, governmental access)Require the operator to take technical, organizational and legal measures against third-country access to non-personal data held in the EU, and confirm them in the contract.Operator attestation of measures, contract clause and configuration evidence for storage location
Data Act Articles 25, 29 and 50 (switching rights from 12 September 2025, no switching charges from 12 January 2027)Design the stack for exit: standard export formats, documented database and object storage layout, and a tested path to another provider or on-premises infrastructure. Confirm the notice and transition terms in each contract.Exit plan, export test results and a current inventory of data formats and dependencies
Backups and disaster recovery (transfer exposure of copies)Keep backups, snapshots and replicas in the EU under the same controls as primary data, including for managed database and object storage services.Backup location record, restore test results and a list of any service providers holding copies
Data Privacy Framework (Implementing Decision 2023/1795, pending Case C-703/25 P)Do not rely on the Data Privacy Framework as the only transfer basis for any US-linked service. Keep a fallback such as standard contractual clauses and a transfer impact assessment, and monitor the appeal.Transfer basis per vendor, dated review record and the fallback mechanism for each US-linked service

Which platforms fit an EU sovereignty brief and why

Two shapes cover most EU sovereignty briefs: an integrated openDesk deployment, or a Nextcloud-centred stack that you assemble from separately licensed components. Both keep data inside infrastructure you control, but they differ in how much integration work you carry, how support is bought, and which licenses you must review.

openDesk as the integrated option

openDesk is a Kubernetes-based suite from ZenDiS GmbH, a federally owned German company that took over development in January 2024. It bundles Nextcloud for files, OX App Suite for groupware, Element for chat, Collabora as the web office, OpenProject, Jitsi, XWiki and Nubus for portal and identity. Each bundled component keeps its own license, for example AGPL-3.0 for Nextcloud and Nubus and MPL-2.0 for Collabora, so legal review must cover each one, not only the deployment repository under Apache-2.0.

Community versus Enterprise editions

ZenDiS recommends openDesk Enterprise Edition for production. It uses the same open-source core licenses but adds closed-source modules, an Enterprise-only registry and license file, and SLA-backed second and third level support and patches. Enterprise features include Nextcloud security patches and the Guard app, Collabora load scaling and OpenProject add-ons. Ask whether your support model depends on these, and whether it is sold as SaaS or for on-premises installation.

Nextcloud-centred stack

A Nextcloud-centred stack assembles Nextcloud for files, Collabora or ONLYOFFICE for documents, Stalwart or mailcow for mail, Mattermost or Matrix Synapse for chat, Jitsi or BigBlueButton for video and OpenProject for projects. You choose each component and its hosting, which gives more control over residency and components but makes you the integrator. Confirm each component's license and support contract before you commit, since they differ across the stack.

Nextcloud Enterprise as a paid option

Nextcloud GmbH sells Nextcloud Enterprise alongside the community version, aimed at enterprise, public sector and education users. Enterprise is a paid subscription with tiers, a user minimum and maintenance terms that differ by tier. Budget it as a recurring cost in the sovereignty case. Confirm which tier includes the patch lifecycle and support hours your operating model needs, and whether the subscription is tied to the number of users.

Major version upgrade cadence

Nextcloud publishes a new major version every four months and supports each for one year. Nextcloud 35 was released on 16 September 2026 and reaches end of life in September 2027. Nextcloud does not allow skipping major releases, so you must move one major version at a time with a backup before each step. openDesk releases minor versions monthly and supports only the latest minor, so plan for a standing upgrade test environment in either stack.

Outline's Business Source License review

Outline is licensed under the Business Source License 1.1, not an open-source license. Its additional use grant forbids using it to offer a commercial Document Service. Before you choose it for a sovereignty brief, have counsel confirm whether your intended use, including use by affiliates or service providers, falls within the grant. For a brief that must be defensible on licensing, this is a review item rather than a default choice.

How sovereign hosting options change operations

Running openDesk or a Nextcloud-centred stack in an EU region of a public cloud gives the organization fast provisioning and managed underlying services, but the organization still owns the workload and its configuration. The sovereignty question shifts from where the server sits to who can reach the control plane, the storage and the logs, and under which law the cloud provider operates. Confirm this with each provider rather than assume it, because EU region placement does not by itself settle access by foreign parent companies.

  • Cluster admin keys: The organization or its implementation partner holds Kubernetes admin credentials. Ask the provider which staff can reach the hypervisor, key management and support tooling, and whether customer-managed keys are available.
  • Backups and object storage: Object storage, database snapshots and backup vaults usually sit in the same tenancy. Confirm the region for each service, including replicas and any cross-region disaster recovery copies.
  • Support path: Provider support is separate from openDesk support. Incidents involving the cluster go to the cloud provider, while application issues go to the suite vendor or partner, so runbooks must name both.
  • Legal exposure: Ask how the provider handles government requests and whether it publishes a transparency process. Map this against the transfer analysis in the EDPB guidance on third-country access.

Before signing, check where support staff are located, which encryption keys the organization controls, and how a move out of the tenancy would work under the Data Act switching rights.

Rollout phases for a sovereign workplace migration

This sequence moves the workplace from a US-controlled productivity cloud to an EU-hosted stack in phases, with a documented output at each gate so the transfer position, identity, data and exit are evidenced rather than assumed.

  1. Assess residency and jurisdiction exposure

    Map every workload, user group and data category that the US service touches, including support access, telemetry, backups and sub-processors. Record where each element is stored and where it can be viewed remotely, because EDPB guidance treats remote access from a third country as a transfer. Review each provider's exposure to third-country authority requests in a transfer-impact assessment, with your legal and data protection team. The output is a data-flow register and a transfer-impact review signed by the data protection officer, with open questions assigned to named owners.

  2. Confirm platform and Kubernetes readiness

    Choose the target stack (openDesk or a Nextcloud-centred design) and check it against the platform requirements your operations team must meet, such as a supported Kubernetes version, ingress, certificate management, storage provisioning and the external databases, caches, object storage, SMTP relay, antivirus and TURN services the chosen product expects in production. Confirm the exact version requirements in the vendor documentation at the time of build, since they change between releases. Decide which team operates each service. The output is a readiness checklist and an operations responsibility matrix, both approved before any production hardware is ordered.

  3. Federate identity to one directory

    Connect the new suite to the directory you already run so users have one identity and one set of access policies across files, mail, chat, wiki and projects. Confirm the federation path for your directory, such as LDAP or Active Directory synchronization or OpenID Connect or SAML federation through an identity provider. Test attribute matching, group mapping and joiner-mover-leaver behavior with sample accounts. Disable local accounts that are not needed for break-glass access. The output is a tested identity design, an attribute mapping document and a log of access reviews run against the pilot groups.

  4. Run a pilot with one user group

    Select a bounded user group whose work is representative but not business-critical, and move their files, calendars, chat and wiki spaces to the new platform. Collect feedback on sharing, co-authoring, mobile access and external collaboration, and log every workflow that depended on a US-only feature. Confirm that the pilot data stays within the agreed boundary by checking the data-flow register against the actual configuration. The output is a pilot report with defect and gap logs, a go or no-go decision recorded by the sponsor, and updated residency documentation for each component in use.

  5. Migrate files and mail

    Migrate files with their permissions, versions and sharing links, and mail with folder structure, flags and calendars, in batches that can be rolled back. Verify checksums or item counts after each batch and keep the source tenant read-only for a defined window agreed in the migration plan. Test mail deliverability, SPF, DKIM and DMARC alignment and retention settings before any user is moved, because coexistence problems usually surface here. The output is a migration reconciliation report per batch, signed by the workload owner, and a record of any items that could not be migrated with the decision taken for each.

  6. Cut over workload by workload

    Move one workload at a time (for example, files first, then chat, then mail) so that each cutover has a single rollback path and a single support queue. Publish a dated cutover notice to users, update DNS and client configuration, and confirm that old links redirect or are reissued. Keep service desk staff briefed on the changed paths. Verify that audit logging, access logging and backup jobs run for the new workload before the old service is switched to read-only. The output is a cutover record per workload with the acceptance checks passed and the rollback test result.

  7. Decommission the US service with a documented exit

    Before terminating the US subscription, confirm that all in-scope data has been exported, verified and deleted from the provider, and obtain written confirmation of deletion, including backups and any support copies, under the terms of your contract. Check retention obligations that apply to records, such as sector rules on message preservation, so that archives are moved to the new platform rather than deleted. Revoke remaining integrations and API tokens, and remove the provider from the data-flow register. The output is a closed exit file containing the deletion evidence, the final register version and the retained-records inventory, reviewed by the data protection officer.

Gaps to plan for before committing to an EU suite

Which features may be weaker than Microsoft 365 or Google Workspace?

Expect gaps in areas such as presence, calendar and meeting integration, AI assistants and admin tooling, and confirm each one with the vendor rather than assuming parity. Open-source components differ in how far they go on these functions, and a bundled suite does not guarantee that every feature your users rely on has an equivalent. Start by listing the daily workflows your staff use, such as shared calendars, co-authoring in Office files, screen sharing in meetings, or delegated mailbox access. Then test each one against the exact component version and edition you plan to deploy. Record the result as a written gap list with an owner and a workaround or acceptance decision for every item. This list is the most useful input for a fair comparison and for the rollout plan.

Which openDesk features require the Enterprise Edition?

ZenDiS recommends openDesk Enterprise Edition for production, and several capabilities are only in that edition. The documented list includes Collabora branding and automatic load scaling, the Element AdminBot, GroupSync and Admin Console, Nextcloud Enterprise security patches and the Guard app, OX S3 storage and advanced full-text search, OpenProject Enterprise add-ons, and XWiki Pro apps. Enterprise Edition uses the same open-source core licenses but adds closed-source modules and SLA-backed second and third level support, sold as SaaS or for on-premises installation. Ask the vendor for the current list, the support response targets in writing, and how an on-premises Enterprise license file is issued and renewed. Budget for it as a recurring cost in your comparison.

Can Nextcloud skip major versions during an upgrade?

No. Nextcloud does not allow skipping major releases, so administrators must move one major version at a time. Each major version is supported for one year through monthly maintenance releases, and the release schedule lists current end-of-life dates for the versions in support. Before every upgrade, take a backup, and disable third-party apps until their compatibility is confirmed. For planning, map the upgrade path from your installed version to the target version against the release calendar, and count how many intermediate upgrades each year will require. Include staff time for testing, app validation and a rollback plan in the operating model. Confirm the exact steps for your deployment method with your implementation partner before go-live.

How often does openDesk release updates, and what does that mean for operations?

openDesk plans major releases annually in Q3, minor releases monthly, and patch releases on demand. Only the latest minor version is supported, in-place upgrades between majors are not guaranteed, and skipping several minor versions may require intermediate steps. This cadence means your team must plan for regular change windows, test each minor release in a non-production environment, and keep the Kubernetes platform, Helm and Helmfile tooling within the versions the release requires. Ask the vendor how security patches are delivered between minor releases, and whether your support model covers testing and rollout of each release. Agree the testing and rollback steps in writing before go-live.

How much retraining will staff need when moving to an open-source suite?

The amount of retraining depends on how different the new interfaces, sharing rules and meeting tools are from what your staff use today, and no general figure should be assumed. Users who change from a Microsoft or Google environment typically need guidance on file sharing links, co-authoring, calendar invitations, and where chat history and wiki content live. Administrators need training on the Kubernetes or server operations, identity mapping, backup and upgrade procedures. Plan the training by role, run a pilot group to measure where users get stuck, and keep help content in the same suite. Ask your partner for training materials and a handover plan, and confirm which of them are included in the written scope.

What should we confirm with each vendor before signing?

Confirm in writing the features your users depend on, the edition and license terms for each component, the supported versions of databases and runtimes, and the upgrade path between the versions you will run. Ask for the support scope, response targets and how security patches are delivered, and confirm whether any component license restricts how you offer it to third parties. For components that are not part of the bundle, such as a separate mail server or a calendar app, confirm the integration method and its maintenance responsibility. Do not rely on a demo or a marketing page as evidence of parity, and ask for reference configurations that match your deployment constraints.

Does DeployOpen handle the gap analysis and rollout planning?

DeployOpen plans, implements, customizes, integrates, hardens, backs up, upgrades, documents and supports self-hosted open-source platforms inside your own infrastructure. Gap analysis and rollout planning can be part of a fixed-scope engagement, where scope, deliverables and timelines are agreed in writing before work starts. Ask for the written scope to name the feature list you need validated, the test cases, the upgrade and rollback procedures, and the training materials that are included. DeployOpen does not publish a price list, so a quote follows from that written scope. Ongoing upgrades and support for the release cadence are available as retainers.

Evidence a regulator or auditor will ask for and which component produces it

This table maps the evidence a data protection authority, auditor or customer security team typically requests to the workplace component or operating process that produces it. Read each row as a question to answer before go-live, then confirm the named source with your own environment and legal counsel. DeployOpen prepares systems, controls and evidence for assessments; independent auditors and certification bodies issue any audit opinion.

Evidence requestedWhere it comes fromWhat to confirm before go-live
Records of processing and data inventoryYour own register, populated from each component's data categories, storage locations and retention settingsWhich personal data each component holds, who the processors are, and which sub-processors sit outside the EEA
Transfer impact assessments for any third-country exposureYour legal assessment, informed by vendor documentation on remote support access and hosting locationsWhether any vendor, support engineer or backup target can reach data from a third country, including display-only remote access
Data-flow diagrams across the suiteArchitecture diagram of Kubernetes namespaces, external PostgreSQL, Redis, object storage, SMTP relay, antivirus and TURN servicesEvery external service is in an EU location you control, and no component calls an endpoint outside the agreed boundary
Access logs and identity audit trailsNubus or your federated identity provider, plus Nextcloud and Mattermost audit logging where enabledLogs are retained centrally, the retention period meets your policy, and administrator actions are traceable to named accounts
Application and infrastructure audit logsKubernetes audit logs, ingress logs and the logging stack you deploy alongside the suiteLog forwarding is configured, logs are tamper-resistant, and someone reviews them on a defined schedule
Backup and restore test recordsBackup tooling for PostgreSQL, MariaDB, object storage and configuration, with dated restore test reportsRestores were tested into a separate environment, the tests ran on the schedule you set, and results were signed off
Vendor and support access logsYour privileged access management records, plus any support session logs from ZenDiS or other vendorsSupport access is time-boxed, approved, recorded, and performed only from locations your data protection terms allow
Upgrade and patch recordsChange tickets, release notes for the component versions you run, and test results from a staging environmentEach upgrade followed the vendor's major and minor rules, with backups taken beforehand and third-party apps checked for compatibility

Which operating model keeps the suite sovereign after go-live

Who should administer a self-hosted workplace suite to keep it EU-only?

Administrators, on-call engineers and support staff with privileged access should be EU-based employees or contractors working from the EU, with access paths that never route through a third-country support queue. Define this in writing before go-live, because a remote administrator in a third country who can view data on a screen can still count as a transfer under the EDPB's Guidelines 05/2021. Confirm with your implementation partner which staff, which locations and which tooling touch production, and make that list part of your records of processing and your vendor onboarding.

How do you control and log vendor access to a sovereign workplace suite?

Vendor access should be time-boxed, approved per ticket, performed through a bastion or jump host inside your network, and recorded. Keep the vendor's accounts disabled by default, require your own identity and multi-factor authentication for every session, and retain session logs in a system the vendor cannot alter. Audit controls of this kind map to the HIPAA Security Rule's audit control requirement in 45 CFR 164.312(b) where ePHI is involved, and they give your compliance team evidence for GDPR accountability. Ask any partner to show the exact access path and log format before signing.

How often do openDesk and Nextcloud releases force upgrades?

Plan around two different cadences. openDesk plans major releases annually in Q3 and minor releases monthly, supports only the latest minor version, and does not guarantee in-place upgrades between majors. Nextcloud publishes a new major version every four months and supports each for one year, and it does not allow skipping major releases. Put both into a standing upgrade calendar with a staging environment, a rehearsal before each production change, and a backup taken before every upgrade. Confirm the current support windows on the release schedule pages before you commit to a window.

Who patches the bundled components, and how fast?

Your operating model must name one owner per component: the suite platform, the Kubernetes layer, the databases, caches, object storage, mail relay and antivirus. For openDesk, the deployment automation is Apache-2.0, but each bundled component keeps its own license, so patch responsibility follows the component and the edition you run. Nextcloud Enterprise adds security patches that the community edition does not provide, and openDesk Enterprise Edition includes SLA-backed patches. Ask any partner for a written patch matrix covering who watches advisories, who tests, who approves and what the target window is for critical fixes.

How do you test exit from a sovereign suite under the Data Act?

The EU Data Act gives customers the right to switch to another provider or to on-premises infrastructure, with a maximum two-month notice period and a 30-day transition, and bans switching charges from 12 January 2027 (Articles 25, 29 and 50). Exit testing means proving you can export files, mail, calendars, chat history, wiki content, project data and identity records in open formats, then import them somewhere else. Run a scheduled export and restore drill at least yearly, record the time and gaps it reveals, and keep the runbook current. Confirm with your provider which formats and tools each component supports for export.

What should you review before engaging a new third-country processor?

Before any new support tool, monitoring service, backup vendor or AI add-on is engaged, assess whether it would transfer personal data outside the EEA, including remote access and cloud storage outside the EEA, which the EDPB treats as transfers. Also assess whether the processor is subject to third-country law with extraterritorial reach, such as the US CLOUD Act at 18 U.S.C. 2713, because complying with a third-country authority request would itself be a transfer. Document the assessment, the data categories, the supplementary measures and the decision owner, and re-run it whenever the processor, its sub-processors or its location changes.

Which skills must stay in-house to keep the suite sovereign?

Keep in-house ownership of identity administration, the decision on who holds privileged access, the upgrade calendar, the exit runbook and the third-country review process. Those decisions carry legal and risk weight and should not sit only with an external party. Operations work such as Kubernetes maintenance, database care and routine patch application can be shared with a partner under a retainer or a fixed-scope agreement. Whether your team can run this alone depends on existing Kubernetes and identity experience, so assess that skill gap honestly before deciding the split.

What recurring costs should you budget for after go-live?

The recurring costs are staff time for operations and administration, Kubernetes and external service hosting, any Enterprise subscriptions per user or per component, retainers for support and upgrades, and periodic security testing and compliance evidence work. Enterprise pricing is set by each vendor: Nextcloud's pricing page lists per-user annual rates starting at 71.29 euros for its Standard tier for 100 or more users, with a 100-user minimum. Ask any partner to split its quote into one-time and recurring lines, state what each retainer covers, and show how the price changes when the user count or the component list changes.

How DeployOpen scopes and runs a sovereign workplace build

DeployOpen plans, implements, customizes, integrates, hardens, backs up, documents and supports self-hosted open-source workplace platforms inside your own cloud accounts, data centers or air-gapped networks. For an EU-hosted suite, that means every component, its data flows and its operating procedures are written down and tested in your environment before you rely on them.

How we scope

  • Inventory which workplace functions move off the US-controlled cloud and which stay for now.
  • Map data residency for each component, including support access, backups, logs and telemetry paths.
  • Agree a written, fixed scope with deliverables, timelines and identity integration boundaries before work starts.
  • Confirm edition and support decisions per component, such as whether a Nextcloud Enterprise or openDesk Enterprise Edition contract is in scope.

How we build

  • Deploy the chosen stack in your infrastructure, using your Kubernetes platform or VM estate as agreed in scope.
  • Connect the suite to your existing directory and identity provider, and test attribute matching with real user groups.
  • Test applications, APIs and infrastructure for security issues and document findings with remediation priorities.
  • Configure logging and audit controls so that the evidence your compliance team needs is produced and retained.

How we hand over

  • Document the architecture, data flows, residency decisions and every external dependency in a handover pack.
  • Deliver runbooks for upgrades, backups and restores, including the one-major-version-at-a-time upgrade path where applicable.
  • Walk your operations, security and compliance teams through the running system and the evidence it produces.
  • Record which controls and evidence DeployOpen prepared, and state clearly that independent auditors issue any assessment outcome.

How we support

  • Offer ongoing support, upgrades and continuous testing as retainers after handover, scoped in writing.
  • Track vendor release and end-of-life dates for each component and plan upgrades against them.
  • Retest the platform after significant changes, such as a major release or a new integration.
  • Agree in the retainer which support hours, response expectations and vendor escalation paths your team depends on.

Sources and further reading

Talk to our team.

Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.

  • 30-minute call: free, with no obligation.
  • NDA on request: we can sign before you share details.
  • Clear next steps: a scope and plan after the call.