Что произойдёт при отказе мастера Redis?

Sentinel зафиксирует сбой кворумом и автоматически повысит реплику до мастера, клиенты с поддержкой Sentinel переключатся сами, остальным достаточно переподключиться.

Пропадут ли данные Redis при перезапуске?

Нет, узлы используют персистентные тома — содержимое переживает перезапуск и пересоздание пода.

Чем managed Redis лучше самосборного Redis на виртуальной машине?

Репликация и автоматический failover через Sentinel работают из коробки, без собственной обвязки и дежурств.

Managed Redis in the cloud — cache with Sentinel and failover.

Managed Redis

Redis for cache, sessions, and queues — deployed as a fault‑tolerant “master + replicas” setup under the supervision of Sentinel. Redis microsecond latencies, but without manual administration of replication and failovers.

In-memory storage and cache with automatic failover

Free Trial

What you get

-1-

Fault‑tolerance out of the box

Master and replicas, automatic switching in case of a failure
-2-

Sentinel monitoring

The cluster status is continuously monitored, and failover occurs without the engineer’s involvement.
-3-

Data integrity

The data is stored on persistent volumes and survives restarts.
-4-

Secure access

The password is generated automatically and stored in secret.
-5-

Standard protocol

Any Redis client works: redis-cli, libraries for all languages.

Redis — не отдельный продукт, а сервис из каталога DevSpace. Он разворачивается внутри вашего пространства и попадает в общий биллинг проекта вместе с остальными сервисами — базами, очередями, кластерами Kubernetes. Приложениям того же пространства кэш доступен по внутреннему адресу, публичный IP для этого не нужен.

Free Trial

How the service works

The service is managed by Redis Operator — an open‑source Kubernetes operator (actively supported development of the well‑known Spotahome operator) that implements the classic production scheme for Redis.

Master + replicas

The write operation is directed to the master, and the data is replicated to the backup nodes. You specify the number of replicas.
Sentinel as an arbitrator

Separate Sentinel processes monitor the health of the nodes. The loss of the master is detected by the quorum, and one of the replicas is automatically promoted — the application only needs to reconnect.
Self-healing

The operator recreates the failed nodes and returns the cluster to the specified configuration without manual intervention.
Persistent volumes

Each node stores data on a persistent volume — cache warming up after a restart does not start from zero.

Opportunities

Use cases

Caching

Reducing the load on the main database, caching pages and API responses.
Sessions and tokens

A fast state storage solution for web applications and API gateways.
Queues and pub/sub

Light task queues and message exchange between services.
Rate limiting and counters

Redis atomic operations for limits, leaderboards, and real‑time counters.

Free Trial

How to deploy Redis in the cloud — three steps

1

Add the “Redis” service to your DevSpace: size, number of replicas, volume size.
2

Take the address and password from the service page.
3

Connect using any Redis client — the protocol is completely standard.

FAQ

Try it out