Redis Cloud Problems & Issues
Real user-reported problems, complaints, and limitations for Redis Cloud. Based on analysis of 10 signals.
Overall score: 70/100
Issues by Category
2024 license change sparked community outrage and fork
In March 2024, Redis abandoned BSD license for SSPL/RSALv2, neither OSI-approved. This 'forced AWS and GCP to create their own fork' (Valkey). Core contributors migrated to forks. Redis 8 (2025) added AGPLv3 option but damage to trust was done.
Memory-based pricing gets expensive at scale
Redis stores data in-memory, and 'having all data in memory can be very expensive.' Large datasets requiring low latency face steep costs. Users report 'difficulty scaling Redis without it becoming prohibitively expensive.' Memory costs compound quickly.
Pro pricing uses hourly billing - unpredictable costs
Redis Cloud Pro starts at $0.881/hour rather than fixed monthly. 'Real-time data workloads spike with launches, shrink when retired, and swing with usage patterns.' Costs can be unpredictable for variable workloads.
Performance over durability trade-off
Redis 'prioritizes performance over durability.' While persistence options exist (RDB, AOF), in-memory nature means data loss risk on failures without proper configuration. Not suitable as primary database for critical data.
Regional cloud provider dependencies
March 2026 AWS me-central-1 zone issue affected Redis Cloud services. 'Timeline for full restoration remains unclear.' Cloud provider outages can impact Redis Cloud availability despite 99.999% SLA target.
ElastiCache locked at Redis 7.2 due to license
AWS ElastiCache 'remains limited to Redis 7.2, preventing access to future features.' The license change means cloud providers won't offer newer Redis versions. Users on AWS must choose between old Redis or Valkey fork.
Valkey lacks Redis 8+ features
Valkey (the AWS-backed fork) 'doesn't support newer Redis versions' and 'lags behind Redis 8 innovations' like hash field expiration, vector search, vector sets. Teams choosing Valkey for licensing miss newer features.
Key-value model limits complex data relationships
Redis uses key-value pairs to store data, making it 'unable to build complex data relationships.' Applications requiring joins, references, or relational queries need different solutions. Not a general-purpose database replacement.
Private offering and contract concerns
Users report 'Redis private offering is not transparent' and 'one-way agreement where customers cannot disengage.' Reports of 'Redis continuing to invoice for unused services' and India support team 'not being collaborative.'
Significant RAM requirements
Redis 'requires significant RAM' to operate effectively. Datasets must fit in memory. As data grows, memory costs scale linearly. Disk-based alternatives may be more cost-effective for larger, less latency-sensitive data.
Regret Patterns
Common reasons users regret choosing Redis Cloud.
Memory costs spiraled as data grew
Teams chose Redis for speed but found 'having all data in memory can be very expensive.' As datasets grew, costs scaled linearly with memory. Should have evaluated whether all data needed sub-ms latency or if disk-based alternatives sufficed for some workloads.
License change disrupted cloud strategy
Organizations using AWS/GCP ElastiCache found Redis versions frozen at 7.2 after license change. New features required either Redis Cloud directly (different billing) or accepting Valkey (missing features). Strategic planning disrupted.
Used Redis as primary database - regretted durability
Teams treated Redis as primary storage, then experienced data concerns when realizing it 'prioritizes performance over durability.' In-memory nature means risk without proper persistence configuration. Should have used traditional database for critical data.
Pro hourly billing exceeded budget expectations
Started with Pro for dedicated infrastructure, but $0.881/hour (~$640/month minimum) plus memory scaling exceeded budget. 'Real-time workloads spike unpredictably' making cost forecasting difficult. Essentials or Upstash would have been more predictable.
Tried to model complex relationships in Redis
Application needed relational data, but Redis is 'unable to build complex data relationships.' Attempted workarounds with multiple keys and manual references became unmaintainable. PostgreSQL or MongoDB would have been correct choice.
When Redis Cloud Breaks Down
Specific scenarios where Redis Cloud fails or underperforms.
Dataset exceeds affordable memory limits
Application data grows beyond what memory-based pricing allows. Costs become prohibitive. Team must implement tiered storage strategy, move cold data elsewhere, or migrate to disk-based alternative for some workloads.
AWS team needs Redis 8 features
ElastiCache stuck at Redis 7.2, but application needs Redis 8 features (vector search, hash expiration). Must migrate to Redis Cloud directly, losing AWS billing consolidation and native integration.
Data loss due to improper persistence configuration
Team didn't configure RDB/AOF persistence correctly. Instance restart or failure loses data. Redis 'prioritizes performance over durability' - proper configuration was required but not understood.
Application needs relational queries
Requirements evolve to need joins, complex aggregations, and relational integrity. Redis key-value model can't support these. Must implement separate database for relational data, adding complexity.
Open-source mandate conflicts with SSPL
Organization policy requires OSI-approved licenses only. Redis 7.4 SSPL doesn't qualify. Must migrate to Valkey (BSD) or wait for Redis 8 AGPLv3. License evaluation should have happened upfront.