You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
**Requirement**: Redis server configured with eviction [`maxmemory-policy`](https://redis.io/topics/lru-cache#eviction-policies):
141
+
**Requirement**: Redis server configured with eviction [`maxmemory-policy`](https://redis.io/docs/latest/develop/reference/eviction/#eviction-policies):
142
142
`volatile-ttl`, `volatile-lru` or `volatile-lfu` (Redis 4.0+).
143
143
Use of LRU or LFU is recommended. It is also possible to use `noeviction`, but it is usually not practical.
144
144
@@ -191,21 +191,21 @@ See `.env`, `config/packages/ezplatform.yaml` and `config/packages/cache_pool/ca
191
191
Persistence cache depends on all involved web servers, each of them seeing the same view of the cache because it's shared among them.
192
192
With that in mind, the following configurations of Redis are possible:
- Shards cache across several instances in order to be able to cache more than memory of one server allows
196
-
- Shard slaves can improve availability, however [they use asynchronous replication](https://redis.io/topics/cluster-tutorial#redis-cluster-consistency-guarantees) so they can't be used for reads
197
-
- Unsupported Redis features that can affect performance: [pipelining](https://github.com/phpredis/phpredis/blob/develop/cluster.markdown#pipelining) and [most multiple key commands](https://github.com/phpredis/phpredis/blob/develop/cluster.markdown#multiple-key-commands)
- Shard slaves can improve availability, however [they use asynchronous replication](https://redis.io/docs/latest/operate/oss_and_stack/management/scaling/#redis-cluster-consistency-guarantees) so they can't be used for reads
197
+
- Unsupported Redis features that can affect performance: [pipelining](https://github.com/phpredis/phpredis/blob/5.3.7/cluster.markdown#pipelining) and [most multiple key commands](https://github.com/phpredis/phpredis/blob/5.3.7/cluster.markdown#multiple-key-commands)
- Provides high availability by providing one or several slaves (ideally 2 slaves or more, e.g. minimum 3 servers), and handle failover
200
-
- [Slaves are asynchronously replicated](https://redis.io/topics/sentinel#fundamental-things-to-know-about-sentinel-before-deploying), so they can't be used for reads
200
+
- [Slaves are asynchronously replicated](https://redis.io/docs/latest/operate/oss_and_stack/management/sentinel/#fundamental-things-to-know-about-sentinel-before-deploying), so they can't be used for reads
201
201
- Typically used with a load balancer (e.g. HAproxy with occasional calls to Redis Sentinel API) in the front in order to only speak to elected master
202
202
- As of v3 you can also configure this [directly on the connection string]([[= symfony_doc =]]/components/cache/adapters/redis_adapter.html#configure-the-connection), **if** you use `Predis` instead of `php-redis`
203
203
204
204
Several cloud providers have managed services that are easier to set up, handle replication and scalability for you, and might perform better. Notable services include:
0 commit comments