Summary
SurrealDB allows bypass of deny-net flags via DNS resolution
SurrealDB offers http functions that can access external network endpoints. A typical, albeit not recommended configuration would be to start SurrealDB with all network connections allowed with the exception of a deny list. For example, surreal start --allow-net --deny-net 10.0.0.0/8 will allow all network connections except to the 10.0.0.0/8 block.
An authenticated user of SurrealDB can use bypass this restriction, using http::<fn>(<url>) functions where the hostname resolves to an IP within the --deny-net block. For example if a SurrealDB administrator wanted to restrict access to other services within a private network and thus set the --deny-net to a network IP range, this could be circumvented by an attacker leveraging DNS records and hostname resolution.
When sending SurrealDB statements containing the http::* functions, if the hostname resolves to a forbidden IP, the SurrealDB server will still issue the request and return the responses to the attacker.
Workarounds
The possibility of this vulnerability being exploited can be reduced by following an allowlist approach to enabling the http capability surreal start --allow-net 10.0.0.0/8 or using the equivalent SURREAL_CAPS_ALLOW_NET environment variable, where endpoints allowed are fully trusted and are not controlled by regular users.
Alternatively, the network access capability can be disabled, using --deny-net or the equivalent SURREAL_CAPS_DENY_NET environment variable without specifying targets, which disables all outbound HTTP, with impact to SurrealDB functionality.
As the impact of this vulnerability depends on the security of the deployment environment of SurrealDB, best practices should be followed within that environment.
Impact
The impact of this vulnerability is circumvention of the --deny-net capability and resulting impact on systems external to SurrealDB. The ultimate impact is dependent on the deployment scenario.
For example, if the SurrealDB server blocks requests to internal/private IP addresses because those services don’t require authentication, but an attacker can still use SurrealDBs ability to resolve their hostnames via DNS and invoke them directly using http::<fn>(<url>), the attacker can access these internal endpoints directly, and potentially retrieve or even alter sensitive information and credentials.
Affected versions
Security releases
Kodem intelligence
Severity tells you how bad this could be in the worst case. It does not tell you whether you are exposed. Exploitability and impact are functions of runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A vulnerable package can sit in your dependency tree and never run.
Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter. Kodem's runtime-powered SCA identifies whether this CVE is reachable in your applications.
Already deployed Kodem?
See it in your environmentNew to Kodem? Get a demo →Remediation advice
A patch has been created that checks resolved hostnames against allowed network targets, preventing http::* functions from connecting to disallowed IPs.
- Versions 2.2.6, 2.3.6 and later are not affected by this issue.
- The first release following 2.1.7 and 3.0.0-alpha.7 and later will not be affected by this issue
Frequently Asked Questions
- What is CVE-2025-71390? CVE-2025-71390 is a medium-severity security vulnerability in SurrealDB (rust), affecting versions >= 2.1.0, <= 2.1.7. It is fixed in 2.1.8, 2.2.6, 3.0.0-alpha.7, 2.3.6.
- Which versions of SurrealDB are affected by CVE-2025-71390? SurrealDB (rust) versions >= 2.1.0, <= 2.1.7 is affected.
- Is there a fix for CVE-2025-71390? Yes. CVE-2025-71390 is fixed in 2.1.8, 2.2.6, 3.0.0-alpha.7, 2.3.6. Upgrade to this version or later.
- Is CVE-2025-71390 exploitable, and should I be worried? Whether CVE-2025-71390 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
- What actually determines whether CVE-2025-71390 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
- How do I fix CVE-2025-71390?
- Upgrade
SurrealDBto 2.1.8 or later - Upgrade
SurrealDBto 2.2.6 or later - Upgrade
SurrealDBto 3.0.0-alpha.7 or later - Upgrade
SurrealDBto 2.3.6 or later
- Upgrade