# OpenFGA v1.11.2

2025-12-04 · [Original release notes](https://github.com/openfga/openfga/releases/tag/v1.11.2) · [Full summary](https://ratatosk.io/en/releases/openfga/v1.11.2)

Security 1 · Important fixes 1

## Security (1)

### Go toolchain upgraded to 1.25.5

- Affects: host certificate validation through `crypto/x509`
- [CVE-2025-61729](https://www.cve.org/CVERecord?id=CVE-2025-61729), [GO-2025-4155](https://osv.dev/vulnerability/GO-2025-4155)

A malicious certificate that makes `HostnameError.Error()` print an unbounded number of hosts could trigger quadratic-time string concatenation and excessive resource consumption; the Go 1.25.5 toolchain includes the fix for this issue.

> - Update toolchain go version to 1.25.5 to address [CVE-2025-61729](https://pkg.go.dev/vuln/GO-2025-4155) in the go std lib.
> - Update toolchain go version to 1.25.5 to address [CVE-2025-61729](https://pkg.go.dev/vuln/GO-2025-4155) in the go std lib.

https://github.com/openfga/openfga/blob/main/CHANGELOG.md

## Important fixes (1)

### `InMemoryCacheController` store-write invalidation fixed

- Affects: setups with `InMemoryCacheController` enabled

After a store write, cached Check responses could remain stale unless multiple Checks happened in rapid succession. Invalidation now uses the last known store modification time but runs asynchronously, so the first Check after a write—and subsequent Checks until the job finishes—can still return a response cached before the write.

> - Fixed an issue with the `InMemoryCacheController` (the default cache controller when enabled) where cached Check responses were not invalidated after a write to the store. Previously, invalidation only occurred if multiple Checks were triggered in rapid succession after a write. [#2811](https://github.com/openfga/openfga/pull/2811)
> - Fixed an issue with the `InMemoryCacheController` (the default cache controller when enabled) where cached Check responses were not invalidated after a write to the store. Previously, invalidation only occurred if multiple Checks were triggered in rapid succession after a write. [#2811](https://github.com/openfga/openfga/pull/2811)

---

Summaries are produced by AI, so they can be wrong, and they are provided without warranty.
