What's new

5.4.117 (54115) - 2024/03/19


  • [Cluster] Avoid logging expected cluster exception
  • [Cluster] Fixed the failover mechanism for cluster transactions
  • [Cluster] Using SemaphoreSlim instead of a blocking Monitor.TryEnter as the cluster lock to prevent potential ThreadPool starvation
  • [Cluster] Fixed an edge case of NRE in ClusterStateMachine when adding a database
  • [ETL] Stopping an ETL tasks in an async fashion to avoid holding the cluster lock
  • [Indexing] Fixed incorrect staleness status due to unprocessed document tombstones after the replication
  • [Indexing] Fixed handling of field names extraction for boosted fields in JavaScript index definition
  • [Licensing] Making sure the license verification check will be executed when running as Windows Service
  • [Revisions] Fixing an issue that Enforce revisions configuration was skipping on docs when passing multiple collections
  • [Subscriptions] Fixed an issue with missing metadata on subscriptions with revisions enabled
  • [Subscriptions] Foxed a task connection status when getting subscription ongoing task by name
  • [ThreadPool] Doubling the minimum workers of ThreadPool in the default configuration
  • [Voron] Fixed occasional failures in loading the storage report
  • [.NET] Updated to .NET 7.0.17


  • Fixed wrong count of dedicated threads on Threads Runtime Info tab
  • Filtering out expired certificates by default

5.4.116 (54109) - 2024/02/12


  • [Backup] Cluster observer will decide which node should perform the backup
  • [Certificates] Avoid triggering the replacement of the server certificate when the new one that comes from the CertificateRenewExec is the same as the one that is currently used
  • [Cluster] Improved performance of cluster operations
  • [Cluster] Avoided materialization of the database record in HandleClusterDatabaseChanged()
  • [Configuration] System store will take into account Storage.Temp configuration option for temporary directory
  • [Debug] Fixed an issue with downloading a memory dump file generated by /admin/debug/dump endpoint
  • [ETL] Replaced deprecated MySql.Data.MySqlClient package by MySqlConnector
  • [ETL] Fixed an issue with not preserving the casing of Time Series names
  • [Import] Fixed an issue that after the import we could end up with two attachments with identical names on one document
  • [Logging] Added ability to persist configuration of the server logs and the traffic watch logging
  • [Querying] Support for nullable numeric parameters of in clause
  • [Revisions] Fixed concurrent usage of the context in the revisions storage
  • [Setup] Fixes in rvn tool for the initial setup operations
  • Fixed issue that a database ID couldn't be added to unused database IDs
  • [.NET] updated to .NET 7.0.15


  • Fixed an issue that WhatChanged() did not return old value for RemovedField on Dictionary
  • Fixed finalization of RequestExecutor instances
  • Throw meaningful exception when trying to use default HiLoIdGenerator while a custom AsyncDocumentIdGenerator convention was set


  • [Security] Introduced TOTP support for browser generated requests

5.4.115 (54106) - 2024/01/18


  • [JavaScript Indexes] Fixed regression introduced in 5.4.113 which made that date fields indexed by JavaScript indexes got different format so queries could not find old results. If an index is affected (has a date field) then the reset is required in order to recover.

  • [Counters] Fixing the allocation size for a counter group name when purging counter tombstones

5.4.114 (54104) - 2024/01/16


  • [Subscriptions] fixed issue with concurrent subscription not picking up appropriate state when connecting

5.4.113 (54101) - 2023/12/27


  • [Attachments] do not throw a concurrency exception on a partial update with valid expected change vector
  • [Backups] fixed issue with restoring subscriptions when snapshot with incrementals is used
  • [Backups] support for Direct Upload for Azure
  • [Backups] added the ability to continue exporting of documents when corrupted data is encountered
  • [Backups] avoid downloading Snapshot file twice in some cases
  • [Cluster] fixed issue with high CPU usage when watcher is waiting for a leader to come up
  • [Cluster] allow to promote rehab node
  • [Cluster] stability improvements when node is moved into rehab and grace period kicks in
  • [Compare Exchange] write result of each cluster command attempt on its own context to avoid using disposed context during failover
  • [Compression] fixed 'End of Central Directory record could not be found' on creation of recovery dictionaries
  • [Compression] fixed issue with using non-compressed schema for a compressed document and the other way around, that could happen if compression is changed in a middle of merged transaction
  • [Compression] fixed issue with missing compression dictionary
  • [Configuration] switched default value of 'Http.AllowResponseCompressionOverHttps' to true
  • [ETL] changed order of elements of Transactional ID used by Kafka ETL producer for easier ACL configuration
  • [ETL] allow to define optional Cloud Event attributes in Queue ETL
  • [Indexes] start tombstone cleaner only after indexes are initialized to avoid cleaning tombstones that were not processed by indexes
  • [Indexes] handle nested reduce result of a reduce key properly
  • [Indexes] properly handle dates stored as JsString in JavaScript Map-Reduce indexes to avoid lowercasing them
  • [Revisions] fixed issue with enforce revisions configuration throwing 'InvalidOperationException' on a collection that does not exist anymore
  • [Revisions] added the ability to adopt orphaned revisions
  • [Revisions] fixed fetching revisions from revision bin, introduced a breaking change due to old API being unusable
  • [Revisions] properly handle out of date revisions when they are not ordered by date and enforce revision configuration is being used
  • [SNMP] added request per second (5s rate)
  • [Subscriptions] fixed issue with concurrent connections not being cleaned up properly
  • [System] added file locker to prevent initializing system store concurrently
  • [Time Series] fixed aggregation with time zone offsets
  • [Unused Database IDs] added server-wide validation

Client API

  • [Changes] introduced a 15s timeout to connect to WS to avoid infinite hangs in some cases when there are network issues
  • [Indexes] handle C# reserved keywords correctly
  • [Querying] throwing an exception when streaming document collection query with includes
  • [Session] fixed possible AVE in WhatChangedFor
  • fixed issue with failover not grabbing the topology from all of the does, which could result in using old topology
  • fixed issue on .NET Framework with DecimalHelper missing field when reflection is used
  • simplified speed check algorithm to perform tests once per minute


  • [Patching] added the ability to disable max statements count
  • added CSRF protection