2026 · 8 versions
Fixed- Fixed a PHP 8.6 deprecation when prefetching
- Fixed prefetching of late-registered non-prefetchable groups
- Fixed groups widget matching unrelated keys in rare scenarios
- Fixed a rare fatal error when using `redis-error.php` drop-in
Changes- Report `noeviction` as an error in diagnostics even when `maxttl` is set
- Escape the cache flush caller in the flush logs
- Prevent CSV injection via cache group names in groups widget downloads
- Hide `tls_options` passphrase and passwords of malformed `url` values in diagnostics
Improved- Update drop-in during single plugin updates and uploads
- Improved support for Relay without in-memory cache
- Improved handling of numeric group names
Fixed- Fixed cache metadata being read from replicas
- Fixed `queryttl` being ignored without `maxttl`
- Fixed `maxttl` being ignored when setting multiple keys
- Fixed stored `null` values being treated as cache misses in some scenarios
- Fixed recursion when the configuration contains an unknown option
- Fixed retention period for metrics snapshots and analytics not being pruned
- Fixed multisite group flushes of numeric groups when the prefix ends with the blog ID
- Fixed site flushes on multisite when `group_flush=full`
- Fixed flushing a non-persistent group flushing the whole cache when `group_flush=full`
- Fixed empty group names flushing every key in site flushes with `network_flush=global`
- Fixed `groupflushlog` missing from the options REST API schema
- Fixed drop-in actions ignoring file modification restrictions
- Fixed `withoutTimeout` not disabling the read timeout on Relay connections
- Fixed bootstrap error details being shown on anonymous admin requests
- Fixed transient cleanup deleting unrelated options
- Fixed latency measurements ignoring credentials and scheme of `servers` URLs
- Fixed `prune-queries` not deleting stale keys, ignoring the prefix and missing keys on clusters
- Fixed analytics returning at most `30` intervals and negative counter deltas after a stats reset
- Fixed fatal error when reading the last error reply on a replicated Relay connection
- Fixed `wp_cache_delete_multiple` not deleting the split `alloptions` hash
- Fixed sorting in the command statistics and slow commands widgets
- Fixed malformed CSV downloads in groups widget
- Fixed cluster connections not being retried in rare scenarios
- Fixed `wp redis cli` ignoring `tls_options`
- Fixed `wp redis watch log|aggregate` edge cases
Changes- Expire stored prefetches after 24 hours
- Support Relay v0.60 and newer endpoint statistics
- Prevent fatal errors while updating to 2.0 beta1
- Support case-sensitive Unix socket paths
- Show network site flushes in the flush log
- Escape cache keys and group names in Tools widgets
- Mask passwords in `servers` URLs in diagnostics
- Require capability for async Site Health checks
1.25.7
5 fixes
Fixed- Fixed cache being flushed when its metadata could not be read
- Fixed mismatched replies after PhpRedis read errors
- Fixed Relay switching to database `0` after connection errors
- Fixed `alloptions` and `notoptions` being read from replicas
- Fixed several `split_alloptions` issues that could leave stale or incomplete options
1.25.6
4 fixes2 changes
Fixed- Fixed metrics missing from `aggregate` and `digest` CLI watchers
- Fixed malformed tags in health check `actions`
- Fixed dashboard widget spacing and overflow
- Fixed some commands being routed to replicas when `cluster_failover=distribute*`
Changes- Warn about using `WP_REDIS_DISABLED` in WP CLI
- Hide configuration constant notice in WP CLI when unnecessary
1.25.5
1 fix3 changes
Fixed- Fixed admin menu update indicator
Changes- Support WordPress 7.x theme colors
- Support Relay's `cluster_(az|distribution_strategy|failover_strategy)` options
- Avoid unnecessary HTTP calls in rare scenarios
Improved- Improved failure resilience of `wp_cache_get_multiple`
- Improved logging of failed transactions
- Improved rare pipeline initialization failure handling
Fixed- Resolved rare deprecation notice
Changes- Preserve valid options in fallback when `Configuration::safelyFrom` fails
- Throw exception in `PhpRedisClusterConnection` when `ping` or `info` fail
1.25.3
1 fix3 changes
Fixed- Fixed Query Monitor 4.x compatibility
Changes- Disable analytics before uninstalling plugin
- Prevent rare fatal error in shutdown handler
- Prevent rare array to string conversion warning
Improved- Improved error handling for `CONFIG` and `SLOWLOG` commands
Fixed- Fixed chart overlapping in WordPress 7.x
Changes- Switch to per-interval deltas instead of cumulative totals in "Requests" charts
- Avoid timeouts by chunking analytics measurements retrieval
2025 · 7 versions
1.25.1
1 new2 fixes5 changes
New- Added optional `analytics.persist_limit` configuration option
Fixed- Fixed purging analytics measurements
- Fixed unit type of Relay "Memory" chart
Changes- Avoid timeouts when flushing with `analytics.persist` enabled
- Limit analytics persist size to `memory_limit / 2` by default
- Retry failed read commands on primary node before switching Sentinels
- Consider `*WithMeta` calls read-only on replicated connections
- Disable cache if `WP_SETUP_CONFIG` is defined
New- Added API/CLI endpoints to reset slowlog and command statistics
Fixed- Fixed accessing `CONFIG` on cluster connections
- Fixed accessing `INFO COMMANDSTATS` on old Redis versions
- Fixed selecting databases on persistent connections
- Fixed KeyDB detection
Removed- Removed transient deletion limit of 10k sites
Changes- Enable `strict` mode by default when a `prefix` is set
- Prevent integrity flush after cache was reset
- Loosen Relay `prefix` health check
- Replaced `current_time('timestamp')` with `time`
- Avoid changing parameter names during inheritance
- Prevent diagnostics warning when `WP_REDIS_DISABLED` is enabled
1.24.5
1 fix
Fixed- Fixed analytics charts not rendering under some conditions
1.24.4
1 new1 fix8 changes
New- Added `objectcache_analytics_snapshot_measurements` filter
Fixed- Fixed `WP_REDIS_DISABLED` not being respected in some places
Changes- Increased adaptive cache configuration defaults
- Support negative `relay.adaptive.events` values
- Switched all `string` diagnostics to `Diagnostic` objects
- Use `#[SensitiveParameter]` for strings that may contain passwords
- Reduce number of measurements fetched for metrics snapshot
- Prevent type error trying to format `null` metric values
- Prevent rare diagnostics error when not connected
- Prevent rare error when uninstalling plugin using WP CLI
Improved- Improved `SLOWLOG` handling on clustered connections
Fixed- Fixed adaptive cache `events` and `ratio` configuration not taking effect
Changes- Support Relay's new adaptive cache `status` property
- Support flushing Relay memory on clustered connections
- Disabled memory usage when scanning groups on clustered connections
- Send sanitized metrics snapshot
- Prevent error when running cron and using foreign object cache
- Prevent error when calling `INFO` and using older PhpRedis versions
- Handle permission errors when running `SLOWLOG|GET` and `MEMORY|USAGE`
1.24.2
1 new1 change
New- Added `wp redis analytics-count` CLI command
Changes- Prevent excessive analytics snapshots
1.24.1
2 fixes4 changes
Fixed- Fixed groups scan API call if permalinks are disabled
- Fixed up some widget padding
Changes- Prevent error when using ElastiCache Serverless instances
- Prevent `unexpected token` errors when scripts are inlined
- Prevent event listener registration when adaptive cache widget is missing
- Switched to ES6 over ES11 syntax
Changelog published by the developer, in their original language.
