#Release Notes

#Compatibility and support matrix

The following table shows the compatibility and support matrix between Alauda Cache Service E1 and ACP versions.

#v5.1.0

#New and Optimized Features

#Cross-Datacenter Replication (Alpha)

Redis 7.2 instances now run a redesigned replication module, offering two modes selected per instance through spec.activeRedis.mode:

• Disaster Recovery (peerof, the default) is the hot-standby topology, with directed links from one upstream to up to 15 downstream instances, each ready to be promoted if the upstream datacenter is lost. It remains available on Redis 6.0.
• Active-Active (mesh) lets every datacenter serve reads and writes locally, with conflicts resolved automatically and membership maintained by a gossip mesh seeded from a handful of addresses. It is available on Redis 7.2 only, and is driven by the new ActiveRedisMesh resource.

Replication and gossip on Redis 7.2 ride the instance proxy, which tells the two transports apart on a single endpoint. Active-Active additionally exposes the module peer port (7379) on the proxy Service, so that mesh members can gossip.

See Cross-Datacenter Replication.

#Mandatory Peer Authentication

Every Redis 7.2 instance with cross-datacenter replication enabled must bind a RedisUser as its peer-auth credential through spec.activeRedis.redisUserRef — normally the instance's own default account, which needs no provisioning. Peer links authenticate with this credential, so every datacenter must carry the same username and password value; a dedicated custom account may be bound instead.

The credential is delivered to the nodes at runtime and is never written into the instance configuration file. Peer links no longer freeze it at wiring time, so rotating the password no longer breaks a link on its next reconnect.

#Upgrade Path from Redis 6.0 to Redis 7.2 for Replicating Instances

An instance with cross-datacenter replication enabled can now be upgraded in place from Redis 6.0 to Redis 7.2. When such an instance carries no peer-auth binding, the platform provisions one automatically, deriving it from the credential the replication group already shares.

The two module generations do not interoperate, so the whole group must be upgraded together within a maintenance window. See Upgrade a Disaster Recovery Group from Redis 6.0 to Redis 7.2.

#Teardown Policy for Replication Links

ActiveRedisConnection now carries a teardownPolicy. Detach (the default) removes the link but keeps the peer's bookkeeping so the link can be re-created later. Decommission permanently retires a peer and releases the retention floors held for it — the correct teardown for a datacenter that is gone for good.

#PVC Backup and Restore on Node-Local Storage

PVC backup no longer requires a ReadWriteMany-capable storage provider. The backup volume is now created with the ReadWriteOnce access mode by default — configurable through spec.accessModes on RedisBackup and RedisClusterBackup — so backups can be stored on node-local block storage. Restore no longer mounts the backup volume into every Redis pod either: the data is delivered to the pods over the network instead of through a shared mount.

Existing backups on ReadWriteMany volumes stay restorable and take the same network-delivered path, and S3 backups keep their existing direct-pull path. The path is chosen by the instance's bundle version, not by the backup's access mode: an instance pinned to a bundle version earlier than 5.1.0 may still perform the previous mount-based restore, and needs its backup on a ReadWriteMany volume.

Selecting a node-local StorageClass in the Web Console additionally requires Alauda Container Platform Data Services Essentials v4.4.1 or later. See Backup & Restore Instance.

#Sentinel Password Enabled by Default in the Web Console

When a Sentinel instance is created — or restored from a backup — in the Web Console, the Connection Configuration section now offers two tabs: Set Redis Password and Set Sentinel Password. On the Set Sentinel Password tab, Set Password is enabled by default and the password fields are required while it stays enabled, so a new Sentinel instance receives a Sentinel credential of its own unless the toggle is deliberately turned off. If you click Create while those fields are empty, the tab label shows a warning icon and creation is blocked.

The Sentinel credential is separate from the Redis credential: it is stored in spec.sentinel.passwordSecret and authenticates the Sentinel nodes, while spec.passwordSecret authenticates the Redis data nodes. Clients must therefore supply the two passwords separately — see Accessing Instances (Sentinel Mode).

Existing instances are not affected. The tabs are only rendered while an instance is being created or restored; updating an instance leaves its Sentinel authentication unchanged.

#New APIs

The ActiveRedis, ActiveRedisMesh, and ActiveRedisInspection resources are documented under Redis APIs.

#Fixed Issues

No issues in this release.

#Known Issues

No issues in this release.

#Security Fixes

No issues in this release.

#v5.0.1

#Fixed Issues

No issues in this release.

#Known Issues

No issues in this release.

#Security Fixes

• Fixed a critical security vulnerability (CVE-2026-31789) in the OpenSSL components (libssl3/libcrypto3) of the underlying Alpine base image. Upgraded the affected packages to patched versions across the Redis server images (6.0, 7.2, 8.4) and the Redis Exporter image.
• Fixed security vulnerabilities in the musl C runtime library (musl-utils), including CVE-2026-40200, in the Redis Tools image. Upgraded the affected Alpine packages to patched versions.
• Fixed a security vulnerability (CVE-2026-25243 / GHSA-c8h9-259x-jff4) involving invalid memory access in the Redis RESTORE command. Upgraded the affected Redis 7.2 and 8.4 images to their respective patched versions (7.2.14 and 8.4.3). Note: The provided Redis 6.0 versions are not affected by this vulnerability.

#v5.0.0

#New and Optimized Features

#Expanded APIs for Redis Modules

The Alauda Cache Service E1 APIs have been expanded to support Redis modules, allowing users to configure and manage modules on Redis instances.

#Fixed Issues

• Fixed an occasional issue where Redis cluster instances would remain stuck in a "processing" state during an upgrade from v6 to v7, caused by a deadlock from conflicting configuration updates.
• Fixed an issue where data loss could occur during the upgrade of Redis Sentinel from v5.0 to v7.2.
  During the upgrade process, data on replica nodes is temporarily cleared. If a pod startup exception occurs during this period, the system fails to perform a proper failover, potentially leading to data loss. This fix optimizes the upgrade logic to ensure data safety and consistency during version upgrades.

#Known Issues

No issues in this release.

#v4.1.0

#New and Optimized Features

#Redis Disaster Recovery Support (Alpha)

Support enabling disaster recovery functionality for Redis instances, including:

• Data synchronization configuration in disaster recovery scenarios.
• Monitoring and alert settings.
• Failover.
• Source-end information management.

#Switching access methods

Support switching between access methods for instances in the Data Services view, including switching between intra-cluster and inter-cluster access methods.

#Fixed Issues

• Previously, when creating a Redis instance in the Data Services view, if the PVC capacity filled in on the UI form was less than the memory capacity, it would cause the instance creation form submission to fail.
  
  This has been fixed in Alauda Cache Service for Redis OSS v4.1.0.

#Known Issues

No issues in this release.

#v4.0.0

#New and Optimized Features

#Component-Platform Decoupling

Alauda Cache Service E1's lifecycle has been decoupled from platform's, which means when customer needs a certain feature or a bug fix, Redis team can release immediately instead of waiting for the platform release together.

#Real-Time Logs on Component Details Page

The logs on Redis details page have been replaced with real-time logs that update instantly, facilitating troubleshooting and issue analysis.

#External Access via LoadBalancer for Redis

LoadBalancer access is now supported for Redis. When creating an instance, users can choose to expose services externally through a LoadBalancer-type Service.

Compare to NodePort, LoadBalancer provides better high availability, performance, and export standard port. It is recommended to use the LoadBalancer method in production environments.

#Limitations

When using LoadBalancer, ensure that the cluster has available external address pools. We recommend using MetalLB to create external address pools for the cluster.

#Fixed Issues

• Previously, during Redis instance upgrades from v6.0 to v7.2, infrastructure instability could (in rare cases) lead to data loss. This issue has been resolved in Alauda Cache Service for Redis OSS v4.

#Known Issues

No issues in this release.