Redis

This release patches multiple vulnerabilities including out-of-bounds writes during RDB loading, use-after-free conditions in TLS pending-data handling, blocked-client processing, and Vector Sets, plus a TLS client certificate authentication bypass caused by embedded NUL byte truncation in the Common Name field. None of these fixes change existing APIs or configuration, so upgrading only requires deploying the patched Redis binary.

Matters for: Operators running Redis with TLS client-certificate authentication, Vector Sets, or RDB files loaded from untrusted or external sources.

Redis 8.8.2 fixes a set of security vulnerabilities including heap out-of-bounds writes and reads in CMSketch, TopK, and Vector Set (HNSW) RDB loading and search paths, use-after-free conditions in TLS pending-data handling and the blocked-client list, an RDB-triggered memory corruption issue that could lead to remote code execution, an ACL key permission bypass affecting SORT, GEORADIUS/GEORADIUSBYMEMBER, and XREAD/XREADGROUP, and a TLS client certificate authentication bypass caused by NUL-byte truncation of Common Names. Upgrading to 8.8.2 requires no configuration changes, but operators should apply the update promptly given the RCE and authentication bypass risks.

Matters for: This matters for teams running Redis with TLS client certificate authentication, ACL-based access control, RDB persistence, or the Vector Sets module in production environments.

Redis 8.10 introduces a compact hash encoding that stores shared field names once per schema to cut memory usage, plus new commands including HIMPORT for bulk hash inserts, LMOVEM/BLMOVEM for multi-element list moves, SUNIONCARD/SDIFFCARD for set cardinality, BACKUP for MP-AOF-based node backup/restore, and new TS.* and FT.* commands for time series and search. It also adds TLS peer certificate-based server-to-server authentication and fixes bugs in blocking commands, AOF loading, replication memory reporting, and RediSearch indexing and aggregation; no existing behavior changes are required to adopt this release.

Matters for: Relevant for teams running large hash-based datasets in Redis who need lower memory footprint, and for operators using RediSearch or Redis time series modules in clustered deployments.