Valkey 8.1.8 → 9.1.0
10 releases · 0 breaking changes · 18 security fixes · 1 advisory fixed
18security0breaking0deprecated3feature134fix
Vulnerabilities fixed by this upgrade
- 7.5 CVE-2026-27623 Pre-Authentication DOS from malformed RESP request
Breaking changes
None in the release notes.
Security changes
- security (CVE-2026-23479) Use-After-Free in unblock client flow in 9.1.0
- security (CVE-2026-25243) Invalid Memory Access in RESTORE command in 9.1.0
- security (CVE-2026-23631) Use-after-free when full sync occurs during a yielding Lua/function execution in 9.1.0
- security GHSA-jcj7-v34w-v9vv: Fix a use-after-free in RDMA connection handling that could allow an authenticated client to crash the server using CLIENT KILL. Only affects servers built with USE_RDMA and configured with an RDMA listener (#4534) in 9.0.6
- security CVE-2026-56684: Fix a use-after-free in TLS connection handling that could allow an authenticated client to achieve remote code execution using CLIENT KILL (#4234) in 9.0.5
- security CVE-2026-63639: Reject corrupt stream RDB files containing a shared NACK across consumers, which could allow remote code execution. Reported by @z0v3r1n and @lifip. (#4073) in 9.0.5
- security (CVE-2026-23479) Use-After-Free in unblock client flow in 9.0.4
- security (CVE-2026-25243) Invalid Memory Access in RESTORE command in 9.0.4
- security (CVE-2026-23631) Use-after-free when full sync occurs during a yielding Lua/function execution in 9.0.4
- security (CVE-2025-67733) RESP Protocol Injection via Lua error_reply in 9.0.3
- security (CVE-2026-21863) Remote DoS with malformed Valkey Cluster bus message in 9.0.3
- security (CVE-2026-27623) Reset request type after handling empty requests in 9.0.3
- security Fixes CVE-2025-67733: RESP Protocol Injection via Lua error_reply in 9.0.2
- security Fixes CVE-2025-67733: RESP Protocol Injection via Lua error_reply in 9.0.0
- security Fixes CVE-2026-21863: Remote DoS with malformed Valkey Cluster bus message in 9.0.0
- security GHSA-jcj7-v34w-v9vv: Fix a use-after-free in RDMA connection handling that could allow an authenticated client to crash the server using CLIENT KILL. Only affects servers built with USE_RDMA and configured with an RDMA listener (#4534) in 8.1.10
- security CVE-2026-56684: Fix a use-after-free in TLS connection handling that could allow an authenticated client to achieve remote code execution using CLIENT KILL (#4234) in 8.1.9
- security CVE-2026-63639: Reject corrupt stream RDB files containing a shared NACK across consumers, which could allow remote code execution. Reported by @z0v3r1n and @lifip. (#4073) in 8.1.9
Every release in between
security(CVE-2026-23479) Use-After-Free in unblock client flow
security(CVE-2026-25243) Invalid Memory Access in RESTORE command
security(CVE-2026-23631) Use-after-free when full sync occurs during a yielding Lua/function execution
featureAdd cluster bus network traffic usage metric in bytes by @hpatro (#3396)
featureReduce latency spikes during rehashing via incremental page release by @chzhoo (#3481)
fixFix(syncio): Set errno on EOF in syncRead and propagate to conn->last by @abmathur-ie (#3580)
fixFix GEOSEARCH BYPOLYGON leak on invalid COUNT by @bandalgomsu (#3568)
fixHandle NULL pointer in streamTrim listpack delta calculation by @smkher (#3591)
fixFixes server crash when RDMA benchmark clients disconnect by @quanyeyang (#3448)
fixFix the memory leak in valkey-benchmark by @nmvk (#3643)
securityGHSA-jcj7-v34w-v9vv: Fix a use-after-free in RDMA connection handling that could allow an authenticated client to crash the server using CLIENT KILL. Only affects servers built with USE_RDMA and configured with an RDMA listener (#4534)
fixFix crashes, assertion failures, and hangs when using RDMA together with IO threads by @quanyeyang (#3335)
fixFix a double free when a module timer callback stops its own timer with ValkeyModule_StopTimer by @quanyeyang (#4211)
fixFix torn RESP3 push frames when a client publishes a large message to a channel it is also subscribed to by @quanyeyang (#4253)
fixRESET now clears CLIENT IMPORT-SOURCE state so reused pooled connections regain normal key expiration semantics by @tjade273 (#3973)
fixTruncated AOF files now discard an incomplete MULTI block entirely, preventing loss of later writes after another restart by @chzhoo (#4342)
fixFix an ACL bypass in GEORADIUS and GEORADIUSBYMEMBER where duplicate STORE options checked only the first destination key by @tjade273 (#3971)
fixFix a use-after-free crash when a cluster message arrives for a message type registered by an unloaded module by @enjoy-binbin (#4360)
fixAlways deep-validate payloads on RDB load and RESTORE, preventing deferred assertion crashes; sanitize-dump-payload is now a deprecated no-op by @jjuleslasarte (#3721)
fixFix out-of-bounds memory access when registering or receiving cluster module messages of type 255, which is now a valid type by @enjoy-binbin (#4410)
fixAOF loading no longer applies ACL checks, preventing silent data loss when replaying commands with a disabled default user by @lukepalmer (#3984)
fixFix a client memory accounting leak that inflated the mem_clients_normal INFO field on replicas after primary disconnects by @enjoy-binbin (#4395)
fixFix a permanent client hang when a blocking command such as BLPOP is pipelined with a partially received next command by @foobar (#4531)
fixHGETEX now requires write permission on the key, so read-only ACL users can no longer change field TTLs or delete fields by @ranshid (#4576)
fixCompare the full TLS certificate CN when authenticating, so an embedded NUL cannot impersonate a truncated ACL username by @madolson (#4577)
fixRestore read performance with IO threads on TCP/TLS by applying extra read-completion handling only to RDMA connections by @quanyeyang (#4414)
fixRestore write performance with IO threads on TCP/TLS by limiting post-write safety checks to RDMA connections by @quanyeyang (#4452)
fixFix atomic slot migration protocol errors with IO threads by not offloading export connection writes while snapshotting by @satheeshaGowda (#4104)
fixReject invalid slot import ranges when loading an RDB, preventing corrupted files from creating bad migration state by @enjoy-binbin (#4229)
fixReject RDB slot-import records with an invalid job name length, fixing an out-of-bounds read during startup by @quanyeyang (#4210)
fixFix a crash when COPY ends with a bare DB token during slot migration, and block cross-DB COPY regardless of option order by @madolson (#4301)
fixHPERSIST, HTTL, HPTTL, HEXPIRETIME, and HPEXPIRETIME now return a syntax error when the FIELDS keyword is missing by @cjx-zar (#4300)
fixFix a TLS and IO threads race that could leave slot migration export jobs stuck until timeout by @jjuleslasarte (#4320)
fixFix a server crash when hash field expirations are set near the maximum timestamp, for example via HPEXPIREAT by @ranshid (#4312)
fixFix a stack overflow crash on TLS connections when retrying a failed write of large replies by @murphyjacob4 (#4307)
fixValidate cluster bus PUBLISH and MODULE packet payload lengths, preventing a remote crash from forged length fields by @tjade273 (#3972)
fixFix a use-after-free crash when serving blocked clients if handling one client frees another blocked on the same key by @quanyeyang (#4212)
fixFix a server panic with IO threads when pipelined commands with a wrong number of arguments reached the key prefetcher by @madolson (#4302)
fixReject crafted stream RESTORE and RDB payloads with inconsistent lengths or negative field counts that could crash the server by @madolson (#3922)
fixReject stream payloads with mismatched live and deleted entry counts that could make XDEL destroy live entries by @roshkhatri (#4381)
fixFix CLUSTER SLOT-STATS ORDERBY returning wrong ordering once cumulative slot counters differ by more than 2^31 by @jzy1688 (#4459)
fixFix atomic slot migration failures with TLS and IO threads by not offloading export connection reads while snapshotting by @satheeshaGowda (#4559)
securityCVE-2026-56684: Fix a use-after-free in TLS connection handling that could allow an authenticated client to achieve remote code execution using CLIENT KILL (#4234)
securityCVE-2026-63639: Reject corrupt stream RDB files containing a shared NACK across consumers, which could allow remote code execution. Reported by @z0v3r1n and @lifip. (#4073)
fixStrictly validate CRLF terminators when parsing the RESP protocol; malformed requests now get a protocol error instead of being misparsed by @enjoy-binbin (#2872)
fixFix a use-after-free crash when creating slot import jobs during manual slot migrations by @twooster (#3283)
fixFix a memory leak in ZDIFF and ZDIFFSTORE when the result set becomes empty before all inputs are processed by @sarthakaggarwal97 (#3342)
fixFix HPERSIST sending a malformed reply that desynchronized the connection when used on a key of the wrong type by @madolson (#3516)
fixFix a crash from a race between IO threads and asynchronous client freeing by @deepakrn (#3458)
fixFix a double free when loading a stream with corrupt consumer PEL data from RDB or RESTORE by @enjoy-binbin (#3498)
fixFix listpack corruption and a subsequent crash when XTRIM marks the last entry of a stream listpack node as deleted by @smkher (#3591)
fixFix malformed replies when module callbacks build deferred-length arrays while a client's deferred reply buffer is active by @eifrah-aws (#3578)
fixFix a NULL pointer crash in TLS pending-data handling by @zuiderkwast (#3641)
fixFix a server crash when multiple RDMA clients disconnect at the same time by @quanyeyang (#3448)
fixFix a use-after-free when ACL LOAD deletes a user whose clients cannot be freed immediately by @ranshid (#3800)
fixFix a use-after-free when a module unregisters the first registered cluster message receiver for a message type by @eifrah-aws (#3846)
fixFix HRANDFIELD looping forever when a hash has fewer non-expired fields than the requested count by @cjx-zar (#4047)
fixFix clients being left on the wrong database after module keyspace notifications for commands like MOVE and COPY by @enjoy-binbin (#4024)
fixFix a Sentinel crash during coordinated failover when the connection to the old primary is disconnected by @lukepalmer (#4068)
fixFix underestimation of client output buffer memory when replies reference shared objects, so buffer limits are enforced correctly by @dvkashapov (#3306)
fixFix a crash on ARM/aarch64 caused by memory-ordering races in the IO thread job queue by @jjuleslasarte (#3878)
fixFix a crash when active hash field expiration leaves a single entry in a large expiration time-bucket by @ranshid (#3950)
fixFix a file descriptor leak when a blocking connection attempt, such as MIGRATE to an unreachable host, times out by @madolson (#3541)
fixFix a potential crash from a dangling slot migration job reference when the migration client is reset by @murphyjacob4 (#3554)
fixRemove cached EVAL scripts when their scripting engine is unregistered, preventing dangling engine references by @eifrah-aws (#3503)
fixFix a memory leak in GEOSEARCH BYPOLYGON when argument parsing fails, such as on an invalid COUNT by @bandalgomsu (#3568)
fixFix a crash when a slot migration target node is removed from the cluster before the migration connects by @chenshi5012 (#3596)
fixFix a crash when the module GetLRU/SetLRU/GetLFU/SetLFU APIs are called with a NULL key by @yaronsananes (#3610)
fixFix an assertion failure in hash field expiration commands when a module blocks the client in a keyspace notification by @enjoy-binbin (#3743)
fixFix a cluster UPDATE log message reading shard IDs past their fixed-length buffer by @enjoy-binbin (#3942)
fixFix undefined behavior in the failover delay calculation when cluster-node-timeout is set below 30 milliseconds by @enjoy-binbin (#3941)
fixReject zipmap RESTORE payloads with overflowing length fields that could cause out-of-bounds access on 32-bit builds by @madolson (#3920)
fixReject NAN scores in listpack- and ziplist-encoded sorted sets on RDB/RESTORE load, preventing a later crash on skiplist conversion by @madolson (#3921)
fixFix a startup crash on 32-bit systems with 64-bit time_t, such as Alpine 3.23, caused by time value formatting mismatches by @chenshi5012 (#3787)
fixFix corrupted client replies when IO threads are enabled, caused by a race between in-flight writes and reply buffer reuse by @nanyan0312 (#4060)
fixCOMMAND INFO in RESP3 now returns the subcommands field as an Array instead of a Set for commands without subcommands by @rickrams (#3939)
fixThe dual-channel replication RDB connection now announces the configured replica-announce-ip, avoiding stale replica entries behind NAT by @jdheyburn (#2846)
fixPrevent replicas from processing stale cluster packets and incorrectly promoting themselves to an empty primary within a shard by @zhijun42 (#2811)
fixSend the replica version on the dual-channel RDB connection so full syncs of data like hash field TTLs no longer fail by @hpatro (#4105)
fixFix slot migration failure handling running twice on ownership changes and an out-of-order error reply in the internal SYNCSLOTS FINISH command by @chx9 (#3723)
fixAllow slot-migration-max-failover-repl-bytes to be set to -1 to disable the limit, as documented by @enjoy-binbin (#3443)
fixFix CONFIG REWRITE producing negative values for memory configs such as maxmemory when set to very large values by @enjoy-binbin (#3440)
+11 more items in the release notessecurity(CVE-2026-23479) Use-After-Free in unblock client flow
security(CVE-2026-25243) Invalid Memory Access in RESTORE command
security(CVE-2026-23631) Use-after-free when full sync occurs during a yielding Lua/function execution
security(CVE-2025-67733) RESP Protocol Injection via Lua error_reply
security(CVE-2026-21863) Remote DoS with malformed Valkey Cluster bus message
security(CVE-2026-27623) Reset request type after handling empty requests
fixAvoids crash during MODULE UNLOAD when ACL rules reference a module command and subcommand (#3160)
fixFix server assert on ACL LOAD when current user loses permission to channels (#3182)
fixFix bug causing no response flush sometimes when IO threads are busy (#3205)
securityFixes CVE-2025-67733: RESP Protocol Injection via Lua error_reply
fixAvoid memory leak of new argv when HEXPIRE commands target only non-exiting fields (#2973)
fixFix HINCRBY and HINCRBYFLOAT to update volatile key tracking (#2974)
fixAvoid empty hash object when HSETEX added no fields (#2998)
fixFix case-sensitive check for the FNX and FXX arguments in HSETEX (#3000)
fixPrevent assertion in active expiration job after a hash with volatile fields is overwritten (#3003, #3007)
fixFix HRANDFIELD to return null response when no field could be found (#3022)
fixFix HEXPIRE to not delete items when validation rules fail and expiration is in the past (#3023, #3048)
fixFix how hash is handling overriding of expired fields overwrite (#3060)
fixHSETEX - Always issue keyspace notifications after validation (#3001)
fixMake zero a valid TTL for hash fields during import mode and data loading (#3006)
fixTrigger prepareCommand on argc change in module command filters (#2945)
fixRestrict TTL from being negative and avoid crash in import-mode (#2944)
fixFix chained replica crash when doing dual channel replication (#2983)
fixSkip slot cache optimization for AOF client to prevent key duplication and data corruption (#3004)
fixFix used_memory_dataset underflow due to miscalculated used_memory_overhead (#3005)
fixAvoid duplicate calculations of network-bytes-out in slot stats with copy-avoidance (#3046)
fixFix XREAD returning error on empty stream with + ID (#2742)
featureTrack reply bytes in I/O threads if commandlog-reply-larger-than is -1 (#3086, #3126). This makes it possible to mitigate a performance regression in 9.0.1 caused by the bug fix #2652.
fixAuthenticate slot migration client on source node to internal user (#2785)
fixBug fix: reset io_last_written on c->buf resize to prevent stale pointers (#2786)
fixSentinel: fix regression requiring "+failover" ACL in failover path (#2780)
fixCluster: Avoid usage of light weight messages to nodes with not ready bidirectional links (#2817)
fixSend duplicate multi meet packet only for node which supports it in mixed clusters (#2840)
fixFix: LTRIM should not call signalModifiedKey when no elements are removed (#2787)
fixFix build on some 32-bit ARM by only using NEON on AArch64 (#2873)
fixFix deadlock in IO-thread shutdown during panic (#2898)
fixFix COMMANDLOG large-reply when using reply copy avoidance (#2652)
fixFix CLUSTER SLOTS crash when called from module timer callback (#2915)
securityFixes CVE-2025-67733: RESP Protocol Injection via Lua error_reply
securityFixes CVE-2026-21863: Remote DoS with malformed Valkey Cluster bus message
fixHSETEX with FXX should not create an object if it does not exist (#2716)
fixFix crash when aborting a slot migration while child snapshot is active (#2721)
fixFix double MOVED reply on unblock at failover (#2734)
fixFix memory leak with CLIENT LIST/KILL duplicate filters (#2362)
fixFix incorrect accounting after completed atomic slot migration (#2749)
fixFix Lua VM crash after FUNCTION FLUSH ASYNC + FUNCTION LOAD (#1826, #2750)
fixFix invalid memory address caused by hashtable shrinking during safe iteration (#2753)
securityGHSA-jcj7-v34w-v9vv: Fix a use-after-free in RDMA connection handling that could allow an authenticated client to crash the server using CLIENT KILL. Only affects servers built with USE_RDMA and configured with an RDMA listener (#4534)
fixFix a double free when a module timer callback stops its own timer with ValkeyModule_StopTimer by @quanyeyang (#4211)
fixRESET now clears the CLIENT IMPORT-SOURCE flag so reused pooled connections stop reading logically expired keys by @tjade273 (#3973)
fixFix AOF recovery of a truncated MULTI/EXEC block that could cause new writes to be lost after a subsequent restart by @chzhoo (#4342)
fixFix an ACL bypass where duplicate STORE/STOREDIST options in GEORADIUS commands escaped key write permission checks by @tjade273 (#3971)
fixFix a use-after-free crash when a cluster message of a module-registered type arrives after the module is unloaded by @enjoy-binbin (#4360)
fixFix out-of-bounds access when registering or receiving cluster module messages of type 255, which is now fully supported by @enjoy-binbin (#4410)
fixSkip ACL permission checks when replaying the AOF, preventing silent data loss when users are restricted or disabled by @lukepalmer (#3984)
fixFix a client memory accounting leak that inflated mem_clients_normal on replicas after disconnecting from the primary by @enjoy-binbin (#4395)
fixAlways validate listpacks on RDB load and RESTORE to prevent deferred crashes; sanitize-dump-payload and the related ACL flags are now deprecated no-ops by @jjuleslasarte (#3721)
fixFix a stack overflow crash when retrying large TLS writes after an OpenSSL write error by @murphyjacob4 (#4307)
fixValidate PUBLISH and MODULE cluster bus packet lengths, preventing a crash from forged packets with oversized payload lengths by @tjade273 (#3972)
fixReject crafted stream RESTORE/RDB payloads with inconsistent lengths or negative field counts that could crash the server by @madolson (#3922)
fixFix a use-after-free crash when serving multiple clients blocked on the same key if one is freed during processing by @quanyeyang (#4212)
fixFix CLUSTER SLOT-STATS ORDERBY sorting when slot statistics differ by more than 2^31 by @jzy1688 (#4459)
securityCVE-2026-56684: Fix a use-after-free in TLS connection handling that could allow an authenticated client to achieve remote code execution using CLIENT KILL (#4234)
securityCVE-2026-63639: Reject corrupt stream RDB files containing a shared NACK across consumers, which could allow remote code execution. Reported by @z0v3r1n and @lifip. (#4073)
fixFix clients being left on the wrong database after module keyspace notifications from commands like MOVE and COPY by @enjoy-binbin (#4024)
fixFix an I/O thread job queue memory-ordering race that could trigger an assertion crash on ARM/aarch64 by @jjuleslasarte (#3878)
fixReject zipmap RESTORE payloads with overflowing length fields that could cause out-of-bounds access on 32-bit builds by @madolson (#3920)
fixReject NAN scores when loading listpack/ziplist-encoded sorted sets, preventing a crash from crafted RESTORE payloads by @madolson (#3921)
fixFix a startup crash when generating INFO output on 32-bit systems where time_t is 64-bit (e.g. Alpine time64) by @chenshi5012 (#3787)
fixFix COMMAND INFO in RESP3 to reply with an empty Array instead of a Set for commands without subcommands by @rickrams (#3939)
fixReject invalid characters in cluster AUX fields and cluster-announce-ip to prevent nodes.conf corruption and injection by @eifrah-aws (#3848)
fixFix lua-enable-insecure-api having no effect when enabled at startup via config file or command line by @enjoy-binbin (#3548)
fixIncrease the maximum process title length from 255 to 1024 characters to avoid truncation with long installation paths by @pkhartsk (#3843)