Released: Dec 18, 2025

December 2025 中的更新

特性

• Security Related Improvements
  • Migrated runtime to Java 24 for enhanced security and long term platform stability.
  • Upgraded third party libraries and removed obsolete ones.
  • Improved encryption handshake protocol, replaced legacy Diffie-Hellman with Elliptic Curve Diffie-Hellman (ECDH) using secp256r1 and added HKDF-based AES key derivation with SHA-256. Simplified supported ciphers to AES/GCM for stronger security and better performance.
• Usability and Administrative Related Improvements
  • Can now set User expiry date.
  • Status text on mouse hover on a status table cell.
  • Local job reports now support admin based syntax.
  • Improved error reporting for task installation/upgrade failure.
  • Engine Authentication now displayed in instance configuration.
  • Jobs can now have a “manual” schedule, meaning they will not run unless manually triggered.
  • Changing to a license that does not include all used functionality will give a warning.
  • Filter field now available in management views.
  • Optional text area in management headers.
  • Selection of instance Worksheet/Management from Monitoring more robust.
  • Now possible to delete the last cloud router access token.
• Performance and Stability Improvements
  • Added caching of hostname property resolution to improve performance.
  • Added optional manual node network cost configuration for mesh routing.
  • Improved Reconnection handling of no engine instances
  • Improved driver handling in instance creation & test connection flows.
  • Introduced version cutoff handling in connection logic.
  • Node network cost calculation implementation to assess node capacity.
• MS SQL Server
  • Performs scheduled backups of selected databases to a shared network backup device, making them available for remote restore by other SQL Server instances.
  • Job which restores selected databases from a shared network backup device created by another SQL Server instance, supporting cross-instance recovery workflows.
  • Enhancements have been made to the management of backup devices within the MS SQL Server platform in the Management section of dbWatch. These improvements were necessary to support the introduction of new dbWatch jobs \(“Push Backup to Devices” and “Pull Backup from Devices”\), where efficient handling of a large number of backup devices and clear visualization of their contents are critical.
  • A new column has been added to the table containing query plans. This enhancement improves performance when retrieving data from the table during analysis in the Management section of dbWatch Monitor, enabling faster and more efficient access to query information.
  • A new context menu has been added in the Management module for the Microsoft SQL Server platform (version 2012 and above), allowing administrators to unlock logins directly from the UI. Additionally, two new menu options have been introduced to configure the `CHECK_POLICY` and `CHECK_EXPIRATION` settings for SQL logins, making login management more accessible and efficient.
  • A new column, `is_copy_only`, has been added to the backup view in Management for Microsoft SQL Server. This allows users to identify whether each backup was created using the COPY_ONLY option. This will allow you to easily distinguish between regular and copy-only backups directly in the table.
  • New farm-level views have been added in dbWatch Monitor to improve visualization and control of database movement across SQL Server instances. These views are designed to support workflows based on the “Push Backup to Shared Devices” and “Pull Backup from Shared Devices” jobs. In environments with multiple SQL Server instances and a high number of databases, the new views provide a structured overview of which databases are being exported or restored, on which instances, and their current status. This enables better operational control, traceability, and insight into large-scale cross-instance recovery and migration processes.
  • This job collects detailed information about currently active SQL sessions by joining data from master.dbo.sysprocesses, master.dbo.sysdatabases, and sys.dm_exec_requests. It extracts session context such as login name, host machine, database, and client program, and correlates it with currently executing SQL statements. The purpose of this job is to complement the “SQL statistics” job, which tracks SQL handle-based performance but does not associate activity with session-level details. This provides more actionable insights into who is executing which queries, and from where. This is particularly useful for troubleshooting or live monitoring.
  • The “SQL Statistics” job has been improved to more accurately retrieve execution plan values and associate them with the correct database. Additionally, enhancements have been made to the visualization of execution plan statistics in the Management section of dbWatch Monitor, providing clearer and more structured insights.
  • Added new parameter exclude_copy_only_databases to the backup check job for AlwaysOn databases. This parameter allows copy-only backups to be ignored when evaluating backup status, ensuring that only regular backups are considered valid for compliance and alerting.
  • Enhanced the filtering of SQL performance statistics in the Performance module for MS SQL Server. Currently, users can sort resource-intensive SQL calls by metrics such as Logical/Physical reads, Logical writes, and Executions. With this improvement, it will be possible to visualize only the SQL calls originating from a specific database, providing more precise insight and easier troubleshooting when monitoring large multi-database environments.
  • Enhanced the SQL performance statistics view in dbWatch Monitor (Management section for MS SQL Server). Individual SQL handle statistics now display all SQL calls grouped by handle ID in a dedicated column. Added charts for worker time and elapsed time to improve visibility and analysis.
  • Collects key SQL Server statistics counters, including Batch Requests/sec, SQL Compilations/sec, and SQL Re-Compilations/sec. This job provides insight into overall query activity and query plan efficiency, serving as a basis for performance monitoring and trend analysis. The enhanced job will provide the data foundation for a new graph in the SQL Performance module within the Management section of dbWatch Monitor.
  • Enhancement to the Management view for MS SQL Server in dbWatch Monitor.
  • When an instance contains a very high number of databases (several thousand), opening the Databases folder in the tree view could previously cause long loading times.
  • Only the first 1000 databases (with dbID < 1001) are initially displayed when expanding the Databases folder.
  • The remaining databases are now grouped into subfolders in chunks of 1000 databases each (e.g., 1001–2000, 2001–3000, etc.).
  • This change significantly improves responsiveness and usability in environments with a large number of databases.
  • The SQL query used to visualize database size metrics in the Management section (graph and details view) has been optimized (MS SQL Server platform). The previous version of the query consumed excessive resources on SQL Server instances with a large number of databases (typically over 1000). This caused unnecessary CPU and I/O load during data retrieval. The new version significantly reduces overhead by improving query logic and limiting expensive operations across large datasets.
  • The conversion procedure from dbWatch 12 to dbWatch Control Center has been improved to include a safeguard: it will now skip the conversion of dbWatch 12 data if the old system has not been used for a long time. This prevents unnecessary or outdated data from being migrated, ensuring a cleaner and more relevant setup in dbWatch Control Center.
  • New parameter “max rows per table” allows dynamic control over how many rows are shown in reports, keeping output concise. Several tables in the report now include clearer descriptions to improve interpretation. The procedure now detects and corrects data inconsistencies caused by migration from dbWatch 12 to Control Center.
  • Log Shipping Monitoring [DEV-43]. The two log shipping monitoring jobs, one for the primary node and one for the secondary node, have been updated with improved support for the ignore databases parameter. You can now use the percent sign (%) as a wildcard character to match multiple database names.
  • Session SQL event collector [DEV-47]. A new job which collects performance statistics from the dbWatchSessionStatsXE Extended Events session. It captures long-running SQL batches and RPC calls along with relevant session and database information. Complements the “SQL statistics” job by adding visibility into who is executing what.
  • New entry views in dbWatch Management provides access to the statistics captured by the Extended Events session and gathered by "Session SQL event collector" job, displaying in a familiar layout for easy comparison and combined analysis.
  • Adds "capture threshold" to optionally persist sub-threshold blocking sessions (recorded with OK severity) for trend analysis, in addition to existing 2 parameters. Default for the new parameter is disabled (0) to preserve current behavior unless explicitly enabled.
  • Added new parameter “exclude copy only databases“ to control whether copy-only backups should be considered when evaluating the last backup time. When enabled, only regular backups are used in the check; copy-only backups are ignored. This applies to both jobs, “Database backup“ and “Database log backup“.
  • Backup all databases [DEV-52]. The reports for the “Backup all databases” and “Backup all transaction logs” jobs have been improved and made more flexible with a new parameter, backup history, which controls how much historical data is displayed in graphs and tables.
  • Improved the existing Lazy Writer and Checkpoint Statistics check by adding collection of the Free List Stalls/sec performance counter. This enhancement will provide deeper insight into buffer pool performance and memory pressure conditions, enabling more accurate monitoring and troubleshooting in dbWatch Monitor.
  • The default job template has been updated to include the new SQL Workload Collector job. This ensures that the collector is available by default in new installations.
  • The comparison incorrectly divided by 1000 instead of 1024, leading to inaccurate evaluations of autogrowth thresholds. The job report has also been updated for clearer indication of which values are expressed in KB and which in MB.
  • Enhanced the visualization of Session Performance data in the Management section of dbWatch Monitor. The goal is to make it easier to identify which programs, hosts, and logins were connected to the SQL instance during a given time period. The updated view should provide clearer filtering and grouping of session data to help users quickly understand connection patterns and activity over time.
  • Implement graphical (experimental) visualization in Farm views to display historical CPU usage aggregated from multiple MS SQL Server instances. This enhancement will allow users to monitor and compare CPU consumption across servers in a single consolidated view.
  • Enhanced the engine script for the MS SQL Server platform to handle invalid or missing default data and log paths retrieved from the Windows registry during dbWatch database creation. If the registry paths are incorrect, the script now catches the exception and creates the database files in the same location as the model database, using its default file paths.
  • Extended timeout settings for the “AlwaysOn transaction log backup” job. The current timeout threshold can cause premature job failure or false alerts.
  • Enhanced the major-version detection properties for Microsoft SQL Server to correctly identify and return version information for SQL Server 2025. This update ensures accurate classification and compatibility across all relevant dbWatch modules that rely on version-based logic.
  • Enhanced the SQL Performance module to better parse and visualize missing index information. Added support for generating recommended index scripts directly from the UI.
  • Enhancement to the blocking session management view: both inst_id and sql_id have been added to the “View last 10 SQLs” context menu option. This provides better traceability and improves troubleshooting of blocking chains.
  • A new monitoring job that detects which “Blitz” procedures (e.g., sp_Blitz, sp_BlitzIndex, sp_BlitzCache) are installed on a SQL Server instance, in which databases they reside, and whether duplicate or inconsistent installations exist. In the Management section, users should be able to view, manage, and optionally execute these procedures. The Farm View should provide an overview of which SQL Server instances have Blitz procedures installed, including version information and any discrepancies.
  • Added a new FDL property named default_windows_domain for the MS SQL Server platform. The property should retrieve and return the default Windows domain name by using the default_domain() function.
  • Introduced a new parameter max databases to the job Index usage statistics (all databases). The parameter defines a threshold for the maximum number of databases on an instance where the job should be executed. The job is primarily designed for environments with a limited number of databases. On systems with a very large number of databases (several thousand), calls to sys.dm_db_index_physical_stats can become too resource-intensive and lead to unnecessary load on the SQL Server instance. By adding this parameter, the job can skip execution when the database count exceeds the specified limit, improving performance and resource control.
  • Develop a new version of the MS SQL Server Patch Status job to include detection of the most recent cumulative updates and security patches (released up to November 2025) for SQL Server 2017, 2019, 2022, and 2025. The updated job should ensure accurate version mapping and improved reporting across all supported SQL Server editions.
• Oracle
  • Updated the compatibility rules for the SYS.AUD$ size job in Oracle. It should no longer be possible to install this job on pluggable databases. Previously, it was incorrectly allowed, which resulted in false alarms when the job was deployed to pluggable databases.
  • The graph that previously displayed only the estimated allocated space for all data files has been improved. It now also includes the actual space used inside the files, so both allocated and used values are shown side by side for better insight.
  • The job now checks the expiration date of the dbWatch database user (for Oracle) and automatically resets the password before it reaches the expiration date. This improvement ensures that monitoring continues without interruption and prevents failures caused by dbWatch being unable to log in to the database.
  • The report has been enhanced with a series of visibility filters that prevent queries from being executed against instances when the underlying objects or tables are missing. Additionally, an initial table has been added to the report to display the status of each instance, providing an early indication of whether the report can be run successfully.
  • Updated the definition and handling of `dbWatch-properties` to improve compatibility with Oracle instances running in “MOUNT“ mode. These improvements ensure that the system no longer attempts to access views or operations that are unavailable in this state and reducing errors in dbWatch log files.
  • Enhanced the exception handling in the “AUD$ Table Size Check” job to make it more robust and stable. The job is now fully compatible with non-pluggable Oracle databases.
  • Improved exception handling in the “Delete SYS.AUD$ data” job and ensured compatibility with non-pluggable Oracle databases.
  • This development task involved improvements across several monitoring jobs related to Oracle Data Guard instances. The updates covered key areas including apply time, sequence apply lag, gap detection, transport status, and startup time, resulting in more accurate and reliable monitoring of Data Guard environments.
  • Engine Compatibility for Selected Oracle Jobs [DEV-34]. This development task ensures that the following Oracle jobs, “Max datafiles”, “Max processes”, “Flash recovery space usage”, and “Datafile status” , are now compatible with instances that have the dbWatch-engine. This prevents them from being placed in the “undefined” group when a standby node becomes primary in a Data Guard environment.
  • A new version (v.3.61) of “Redo statistics“ job, which now includes, among other improvements, a report section that shows counts of redo log switches per day. Each row represents a specific hour and the number of switches recorded during that hour. The table is used to analyze daily and hourly trends in redo log activity.
  • The job “Dataguard Archive Sequence Apply Lag”, intended to run on physical standby databases, has been improved to better handle non-standby environments. If executed on a database that is not in standby role, the job will now clearly report the current “database role” and will no longer raise a warning or alarm status in such cases. The job is renamed to “Dataguard archivelog apply gap”.
  • Implemented new FDL properties for Oracle sessions, including sub-properties such as machine, logon_time, suser, status, program, and others. These properties should only apply to the Oracle platform and will enhance session-level monitoring and diagnostics within FDL.
  • The conversion procedure from dbWatch 12 to dbWatch Control Center has been improved to include a safeguard: it will now skip the conversion of dbWatch 12 data if the old system has not been used for a long time. This prevents unnecessary or outdated data from being migrated, ensuring a cleaner and more relevant setup in dbWatch Control Center.
  • New properties are needed for Oracle CDB and PDB databases to determine the correct target location for the dbWatch schema specification after a switchover or failover event. These properties should allow dbWatch to identify which container and pluggable database currently host the active schema and ensure automatic relocation or validation of the schema specification during role transitions in an Oracle Data Guard or RAC environment.
  • Improved Output for Primary Instances in “Replication Delay Alert” Job [DEV-160]. Enhancement of the Replication Delay Alert job to provide clearer and more complete output on primary instances.
  • Database uptime calculation in Oracle DataGuard environments [DEV-141]. Property definitions related to database uptime calculation have been reviewed and improved for Oracle DataGuard environments. The previous logic did not consistently account for role transitions (Primary – Standby) or temporary disconnections in DataGuard setups, which could result in inaccurate uptime reporting.
• PostgreSQL
  • Added the standby arrow to Patroni Secondary instances.
  • Three dbWatch jobs which monitor the presence and freshness of full, differential, and incremental pgBackRest backups. Each job checks that at least one recent backup of the respective type exists, ensuring a complete and recoverable backup chain. The jobs rely on an external script that extracts backup metadata using ‘pgBackRest info —output=json’ and inserts it into a PostgreSQL table. This table does not need to reside in the `dbwatch` schema but must be accessible within the dbWatch monitored database.
• Azure DB
  • The “Rebuild Indexes” maintenance job for Azure single database instances has been updated with a new parameter: ‘rebuild_timeout’. This parameter defines the maximum number of seconds allowed for an individual index rebuild operation before it is automatically terminated.
• Azure
  • The three Azure performance monitoring jobs, CPU, IO, and Log Write, have been updated with a new parameter: retry_threshold. This parameter controls how many times the job will retry fetching data before reporting an error. The enhancement helps prevent false alarms caused by transient connectivity issues or external Azure-related delays, ensuring more robust and reliable monitoring.
• DBW
  • Updated the “DB count” column in dbWatch Monitor by adjusting its width for better readability and changing the column title to ensure consistent naming and visual alignment with other columns in the view.
  • Enhanced the visualization of offline instances in dbWatch Monitor. When an instance is offline, only the information available from the dbWatch Server side is displayed. This provides a clearer and more accurate representation of instance status and available data.
  • Introduced a new dbWatch FDL property instance_class for all supported database platforms (Oracle, SQL Server, PostgreSQL, MySQL, etc.). The property returns a concise, human-readable classification of the connected SQL instance’s architecture/topology and role (e.g., cluster/HA type and primary/secondary state). The value will also be used to auto-render an additional icon in dbWatch Monitor where appropriate.
  • Enhanced the Internal View to improve clarity and usability when managing multi-domain dbWatch environments. The changes include small layout and visualization adjustments to make it easier to distinguish and navigate between domains and servers.

修補程式

• Security
  • Cloud router access token no longer plaintext and has an optional “Hint” field.
  • Worksheet authentication flushing issue addressed.
• Usability and Administrative
  • “Copy to clipboard” menu added when instance connection errors occur.
  • Global component load time display.
  • Syntax verification before resource upload.
  • Status text on mouse hover on a status table cell.
  • Deletion of a security group/role while in use is no longer allowed.
  • Assigning an exhausted license to an instance no longer possible.
  • Engine Authentication now displayed in instance configuration.
  • Jobs can now have a “manual” schedule, meaning they will not run unless manually triggered.
  • Changing to a license that does not include all used functionality will give a warning.
  • Instance configuration “Save Changes” question less annoying.
  • Optional text area in management headers
  • Resolved issue with incorrect OK reporting when testing access points.
  • Self deletion of logged in user blocked.
  • Fix in offline task topic handling when system is in offline mode.
  • Fixed bug when handling job parameters overrides in templates.
  • Fixed bug sorting of view tabs in management.
• Performance and Stability
  • Installer migrated from Install4j to MSI for Windows platform.
  • Bug fix when handling multiple versions of the same task in a resource bundle.
• Minor Bugfixes
  • LineCharts can now handle dates (and not just datetime) on the time axis.
  • LineChart did not handle resultsets where not all columns were plotted.
  • Compatibility resolving in Monitoring tabs.
  • Handling of name/role collisions in the Domain Configuration dialog.
  • Fixed bug in report layout that caused charts not to load.
  • Fixed “header¬icon misalignment” in management filters.
  • Bug fix for special characters in regex handling of FDL-like function.
  • Handling of dynamic values in management dialogs.
  • Bug when handling job parameters overrides in templates.
  • Bug fix for sorting of view tabs in management.
  • Bug fix for Copy cell with large cell content.
• MS SQL Server
  • Fixed an issue in the Farm View module under Security and Compliance where the table “MS SQL Server instances with no [job-name] issue” previously failed to display data. The table now correctly lists all instances that are compliant and have no issues detected by the specified job.
  • A bug was identified in the SQL Statistics job where some performance statistics entries were missing the associated SQL text. This issue has now been resolved, and the job should be updated to include the fix in future releases. The root cause was related to incomplete handling of specific plan or cache scenarios. A new column was also added to the main table that shows the first 1000 characters of an SQL text.
  • Resolved an issue where the property ms_inst_names used incorrect FDL syntax in the filter part of the FDL query, causing the query to fail.
  • Fixed an issue in the “Restore database” menu for the MS SQL Server platform where the default value for the backup device was missing. The menu now correctly preselects the default backup device to improve usability and prevent restore configuration errors.
  • Bug fix in the Autogrowth Settings job procedure where parameter values (expressed in MB) were compared with storage values for database files (expressed in KB).
  • Fixed permission validation in Management menus.
  • Fixed bug in the Scheduler view where selecting “Show jobs using schedule” always returns “No data”, even when jobs are correctly linked to the schedule. The issue occurs due to conversion errors in columns such as next_run_date and next_run_time, which are stored as integers in SQL Server and are not properly converted during view rendering.
• Oracle
  • The dbWatch job “Applied Archive Log Gap Status” was updated to handle a specific error: numeric or value error: character to number conversion error. The fix ensures more robust parsing of input values. By default, the job analyzes only active archive log destinations, in line with standard Oracle configuration practices.
  • This development task addressed an issue where, in pluggable Oracle databases, rows from the alert log were being read multiple times. This could lead to the same error message being triggered repeatedly. The logic has been corrected to ensure that each alert entry is processed only once, improving accuracy and reducing noise in monitoring.
• Azure
  • The Azure database job responsible for gathering DML statistics had an error that prevented the test from being triggered. The issue has been corrected, and the job now runs and triggers the test as expected.