Security at Rehive
How we protect the Rehive platform and your data, what we are certified for today, and how to report a security issue.
On this page
Rehive runs the platform and apps that our clients use to launch and operate their own financial products. Our clients' programs move real money, so security is built into how we design, operate and support the platform.
Each section below describes the controls we have in place today. We keep strengthening them and update this page as new controls go live. If your security review needs something that is not here, email [email protected].
Compliance
| Area | Status |
|---|---|
| SOC 2 Type II (Rehive) | In progress We are preparing for a SOC 2 Type II audit. The report is targeted for mid-2027. |
| CSA STAR Level 1 (CAIQ) | In progress We will publish our completed Consensus Assessments Initiative Questionnaire on the CSA STAR registry. |
| Independent penetration test | In progress A third-party penetration test of the platform. A summary will be available on request. |
| Infrastructure (Google Cloud) | Certified The platform runs on Google Cloud, which holds SOC 1, SOC 2, SOC 3, ISO/IEC 27001, 27017 and 27018 certifications, among others. |
Until our own SOC 2 report is available, Google Cloud's reports cover the infrastructure layer the platform runs on:
- Google Cloud SOC 2
- Google Cloud ISO/IEC 27001, 27017 and 27018
- Google Cloud compliance reports, where Google publishes its audit reports
Policies and people
- Written security policies. Our security practices are set out in written policies, including vulnerability and patch management, and incident response, business continuity and disaster recovery. Each policy has a named owner, is approved by our CTO and is reviewed at least once a year.
Identity and access
- Phishing-resistant sign-in. Every Rehive workforce account is enrolled in Google's Advanced Protection Program and signs in with a hardware security key.
- Dedicated privileged accounts. Privileged work on our cloud infrastructure is done from dedicated admin accounts, separate from everyday accounts.
- Permissions managed as code. Cloud access permissions are defined in code and changed through reviewed pull requests, so every change has an author, a reviewer and a history.
- Least privilege in production. Access to production Kubernetes is scoped by role to the services each team runs.
- Break-glass access. A separate break-glass path exists for incidents, and every use is logged.
Security for your program
The controls your own team uses are the same for every program and cannot be weakened:
- Multi-factor authentication for staff. MFA is always on for every admin and staff account on your project, and cannot be switched off.
- Second factor for users. Every user sets up a second factor during onboarding. Sending money, and adding or removing an email address or mobile number, is confirmed with a one-time code.
- Role-based permissions. Staff accounts are either Admins, who manage your project, or Support, who can view users and transactions without creating transactions or changing configuration.
- Limited Rehive access. Rehive has one review account on your project, used only to support you and to apply platform controls. We never hold or move your users' funds.
See Sign-in and security, Team and staff access and Rehive access to your project for the details.
Secrets and key management
- Encrypted secrets. Kubernetes secrets are encrypted a second time, on top of Google's default encryption, with keys we manage in Cloud KMS, rotated automatically every 90 days.
- Passwords. We store only hashes of user passwords, made with Argon2, the algorithm OWASP recommends.
Infrastructure and deployment
The Rehive platform runs on Google Cloud in the europe-west4 region (Netherlands), on Google Kubernetes Engine.
- Spread across three zones. Production runs on a regional Kubernetes cluster with nodes in three separate zones (data centers), so losing one zone does not take the platform down.
- Private by default. Servers have no public IP addresses, and the cluster's control plane has no public endpoint. Inbound traffic reaches the platform only through Google's load balancers, and outbound traffic leaves through a managed NAT gateway.
- Hardened, patched nodes. Nodes run Google's Container-Optimized OS with Secure Boot and integrity monitoring, and Google applies Kubernetes and operating-system security updates automatically.
- Edge protection. Our public domains are served through Cloudflare, with a web application firewall (the OWASP Core Ruleset and Cloudflare's managed rules) and DDoS mitigation. Google Cloud Armor adds adaptive Layer 7 DDoS protection at the load balancer.
- Separate environments. Staging and production run on separate Kubernetes clusters.
Vulnerability management
- Container scanning. Every container image in our registry is scanned for known vulnerabilities when it is pushed, and rescanned as new vulnerabilities are published.
- Dependency monitoring. GitHub Dependabot monitors our code repositories for dependencies with known vulnerabilities.
- Severity-based fixes. An engineer triages every critical and high-severity finding and sets its severity for how Rehive actually uses the affected code. Fixes are deployed within set timelines for each severity, and any exception needs the CTO's approval.
- Managed platform patching. Google applies Kubernetes, node and database security updates automatically, within our maintenance window.
Data protection
- Encryption in transit. Our public endpoints accept only TLS 1.2 and TLS 1.3. Traffic inside Google's network is protected by Google's encryption in transit, and connections to our databases are encrypted.
- Encryption at rest. All data is encrypted before it is written to disk, with Google Cloud's default encryption.
- Databases are not on the internet. Production databases have no public IP address and are reachable only from our private network.
- Backups. Databases are backed up daily, and transaction logs are kept so a database can be restored to any point in the last seven days. Backups are encrypted and stored in Google Cloud's EU multi-region.
- Data location. Production databases and their backups are stored in the European Union.
Our Privacy Policy explains how we handle personal data.
Availability and continuity
- Uptime goal. Our internal goal is monthly uptime above 99.95% for the Rehive platform, extensions and apps. You can follow live status and past incidents at status.rehive.com. Enterprise clients can have service levels set in their agreement.
- Resilience and recovery. Production databases have a standby in a second zone. If a zone fails, they fail over automatically, typically within minutes and without data loss. For incidents that need a restore, such as data corruption, our recovery objectives are a recovery point (RPO) of under 15 minutes and a recovery time (RTO) of under 12 hours. Backups are held in Google Cloud's EU multi-region, so the platform can be rebuilt in another EU region if needed.
- Continuity plan. A written business continuity and disaster recovery plan covers zone failure, faulty releases, data corruption, compromised accounts and loss of a cloud region.
- Maintenance. We upgrade without downtime wherever we can. Work that might cause a brief interruption is done in a maintenance window, Tuesday to Thursday, 12:00 to 18:00 UTC. Most maintenance causes no interruption, and any interruption is typically brief. If we expect maintenance to cause more than 15 minutes of downtime, we tell clients at least seven days ahead, by email or Slack.
Application security
- Injection. Database queries are parameterised: the query is defined separately from its values, which the database driver escapes.
- Cross-site request forgery. API requests authenticate with bearer tokens, and cookie-based sessions are protected against CSRF.
- Cross-site scripting. Our web apps encode user-supplied content by default before rendering it.
Logging and monitoring
- Audit logs. Administrative actions on our cloud infrastructure are logged by Google Cloud and kept for more than a year.
- Sensitive access is logged. Reads of secrets held in Google Secret Manager, and access to identity and permission systems, are recorded in data access logs.
- Platform monitoring. Application errors and production alerts go to an on-call engineer, and incidents are published on status.rehive.com.
Vendors and subprocessors
The platform runs on Google Cloud, with Cloudflare at the edge. The full list of providers that process data on our behalf, with what each one does and where, is on our subprocessors page.
Incident response
- A documented plan. We follow a written incident response plan with named roles, severity levels and an on-call rotation. Serious incidents get a written post-incident review, and follow-up actions are tracked to completion.
- Telling you. If a security incident affects your data, we notify you within 24 hours of confirming it, by email and in your shared Slack channel, and keep you updated until it is resolved. Platform outages are posted on status.rehive.com.
- Supporting your obligations. You remain the controller of your users' data. We give you the facts and records you need for any notice you must send to your users or regulators.
If you think your own account or keys are compromised, see Security incidents for what to do straight away. For urgent issues on a live program, use emergency support.
Report a vulnerability
We welcome reports from security researchers. Email [email protected]. What a report needs, what is in scope and the rules for testing are in our vulnerability disclosure policy.