Changelog (2024)¶

December 25, 2024¶

OpenAPI Updates¶

1. Resource Catalog: Supports obtaining the Resource Catalog list, supports Create/Modify/Delete/Export Resource Explorers.
2. Generate Metrics: Supports listing the Metrics list, supports Create/Modify/Enable/Disable/Delete Metrics.

Feature Updates¶

Pipeline¶

Automated Pipeline Writing: Traditional log parsing methods often rely on manual script writing based on familiarity with rules, which not only consumes a lot of time but is also prone to errors, leading to inefficient Pipeline writing and inability to respond and resolve issues in a timely manner. To improve efficiency and reduce labor costs, this feature leverages the natural language processing capabilities of large models to assist in automated log content parsing, helping developers extract valuable information from large volumes of logs more efficiently and accurately, simplifying data flow and processing.

Monitoring¶

AI Alert Compression and Merging: Large models are used to compress and merge monitoring alert events, reducing redundancy in alert events, especially when the system generates a large number of alerts in a short period. By integrating alert events, it ensures that Ops personnel receive only one comprehensive alert within a specified time period (e.g., 10 minutes) instead of multiple duplicate alerts.

Cloud Billing Analysis¶

Cloud Billing scenario views and explorers now display consumption currency categories.

Scenarios¶

1. Dashboards and built-in views now include Identifier ID: Dashboards and user views can be uniquely identified by their Identifier ID, and chart configuration jump links can locate dashboards or user views using the Identifier ID.
2. Explorer Auto Refresh logic adjustment: The "Pause" button has been removed from the time component, and the refresh frequency options now include "off," allowing users to turn off auto-refresh as needed, making operations more concise and intuitive. Additionally, the "Explorer Auto Refresh" switch in User Settings has been removed to avoid redundancy and optimize the user setting experience.
3. Explorer now includes the MD5() function, supporting querying log data with MD5 encryption as the desensitization method.
4. Chart optimization: Charts now support obtaining PromQL metric names as legends, and {__name__} can be used to configure metric names as aliases.
5. PromQL editor now includes syntax highlighting and syntax hints.

Metrics¶

Generate Metrics optimization:

1. All types of generated metrics are integrated into the Metrics Catalog, making it easier for users to view and manage them uniformly.
2. Supports re-aggregation of existing metric data to generate new metrics, providing more dimensions for data analysis.

Management¶

Data access and sensitive data viewing desensitization rule application optimization: Optimized the combination of multiple rules and roles to determine data viewing scope. If a member has multiple roles, the final data access scope is the union of the scopes defined in the rules. The same applies to sensitive data viewing desensitization rules.

Infrastructure¶

Resource Catalog Cross-Workspace Query: The Resource Catalog menu now supports cross-workspace querying of resource list data.

Deployment Plan Updates¶

1. Deployment Plan supports Custom Navigation Bar and Action Button Display Text.
2. Management backend member menu optimization:
   • Added a "Login Type" column to distinguish between default login members and SSO login members, and supports filtering by login type;
   • For SSO login members, supports editing the "Other Attributes" field, allowing custom attribute information. Such information is synchronized to Webhook notifications and the corresponding event content area when sending alerts.
3. For workspaces with Volcengine as the primary storage engine, the storage strategies for traces, logs, and custom log indices are adjusted to: Standard Storage, Low-Frequency Storage, and Archive Storage, with configurable durations for hot data, low-frequency data, and archive data storage. Additionally, if the primary storage engine is Volcengine, the storage strategy for new indices in the TrueWatch console is also adjusted to: Standard Storage, Low-Frequency Storage, and Archive Storage.

New Integrations¶

• Added Volcengine ALB integration;
• Added Volcengine CLB integration;
• Added Volcengine NAS integration;
• Added Kube State Metrics integration;
• Added Ranger admin integration;
• Added Ranger tagsync integration;
• Added Ranger usersync integration;
• Updated Redis dashboards and monitors;
• Updated nginx integration & dashboard;
• Updated RabbitMQ dashboards and monitors.

Bug Fixes¶

1. Fixed the issue where host tags did not take effect;
2. Fixed the issue where updating tokens in the console failed;
3. Fixed the issue where sorting in composite dashboards was lost after refresh operations;
4. Fixed the issue where table charts in shared dashboards did not display data in edit mode;
5. Fixed the issue where switching between different charts in cross-space authorized query dashboards did not trigger backend requests;
6. Fixed the issue where version data in RUM views was abnormal;
7. Fixed the issue where built-in Func prompts in monitors could not find the account and reported a 404 error without generating events;
8. Fixed the issue where the user access monitoring list displayed 0 data;
9. Fixed the UI display issue at the bottom of the AI Assistant page;
10. Fixed the issue where a resource classification in the Resource Catalog still existed after successful deletion.
11. Fixed the issue of false alerts due to data gaps.

December 11, 2024¶

Feature Updates¶

Scenarios¶

1. Topology Map now supports external data queries: Users can use DataFlux Func to draw topology maps with external data. Users only need to connect data according to the chart structure to easily visualize external data.
2. Chart optimization
   • Chart display optimization: The chart display has been optimized to slide out from the side, categorized for easier finding and use.
   • Chart descriptions and use case displays: Chart lists now include descriptions and use cases to help users better understand and select the appropriate chart types.
3. View variable optimization
   • View variables now support configuring value listing limits to avoid performance issues caused by too much data listing.
   • The default listing limit for the Deployment Plan is 50, and supports custom default listing quantities. Note: Page configuration limits take precedence over global limits.

Infrastructure¶

1. Resource Catalog optimization:

   • Added grouping functionality: To improve resource management efficiency, resource grouping functionality has been added. Users can group resource classifications with commonalities for easier management and viewing, optimizing resource organization.
   • Explorer honeycomb chart mode optimization: Supports configuring "Color Fill" and "Group Analysis" field lists, allowing users to customize optional fields as needed.

2. Container and Resource Catalog explorer search optimization: Container explorer now supports container_name search, and Resource Catalog now supports name search.

Monitoring¶

Alert strategy configuration optimization

• Supports one-click creation of associated monitors through the alert strategy entry, adding a unified alert notification management mode centered on alert strategies.
• Optimized the display interaction of the alert strategy list association.

RUM¶

RUM now supports React Native application types and allows uploading SourceMap for data parsing and restoration. If you need to upload SourceMap for React Native applications originally created under Android / iOS types, please note:

1. When creating a new application, select the "React Native" application type. During creation, you can directly copy the name and application ID of the existing React Native application under the Android / iOS type, then click the "Create" button. After creation, you can configure SourceMap under this application. This change only affects the creation method of React Native applications and does not affect data collection.

2. If the React Native application under the original Android / iOS type has associated monitors and has not been recreated under the "React Native" type with the same ID, the original monitors can continue to work normally. However, if you have already recreated the "React Native" type application, the original monitors will not be able to obtain data under the "Application Name" item and will display as empty. In this case, you need to select the application type as "React Native" and resave the monitor.

APM¶

APM installation guide now includes automatic injection method: The APM (Application Performance Monitoring) installation guide now includes an automatic injection installation method using Kubernetes Operator. This method simplifies the APM deployment process, allowing users to install and use APM more quickly in Kubernetes environments.

New Integrations¶

• Added HBASE region integration;
• Added HBASE master integration;
• Optimized NodeExporter integration view;
• Added Huawei Cloud DCAAS Cloud Direct Connect integration;
• K8S dashboard adjustment;
• Updated memcached view and monitor;
• Updated rabbitmq view and monitor.

Bug Fixes¶

1. Fixed the issue where expression query values were abnormal;
2. Fixed the issue where querying the CPU usage rate of pod started at a certain time using PromQL in a chart showed data within the time range in a time series chart but did not show data when converted to a chart query.
3. Fixed the issue where alert events did not recover.
4. Fixed the issue where PromQL query results were abnormal.
5. Fixed the issue where modifying hot storage duration in private deployment > management backend reported an error due to the warmretention field missing.
6. Fixed the issue where log search results in the Deployment Plan were inconsistent with searches on Volcengine TLS.
7. Fixed the issue where embedded dashboards in user IFrame pages would randomly scale.
8. Fixed the issue where creating a new group in user management and adding it to multiple spaces caused page lag.
9. Fixed the issue where entering the TrueWatch console > events reported an error with df_fault_id.
10. Fixed the issue where opening the events menu would prompt exceeding memory limits.
11. Fixed the issue where overview chart expression calculations were incorrect.
12. Fixed the issue where monitors triggered alerts but did not generate events.
13. Fixed the issue where the log query function in the Deployment Plan Volcengine base was abnormal.

November 27, 2024¶

Breaking Changes¶

1. OpenAPI: If configuring alert strategies via API to notify members by rule mode, note that the parameter structure for adding/modifying member-type alert strategies via OpenAPI has been adjusted.

2. Dashboard > Visibility Range: Added a "Custom" option to configure operation and view permissions for members of this dashboard.

   • Note: If you previously added "Teams" to the visibility range, team configurations will become invalid and need to be reconfigured.

Feature Updates¶

Cloud Billing¶

1. Added feature guide page: Provides clear steps and explanations to help users get started quickly;
2. Added support for Volcengine and Microsoft Cloud data access.

External Data Source Access¶

1. MySQL data storage system support: The platform now supports accessing MySQL data storage systems, allowing users to perform real-time data queries and analysis.

2. Native query statements can be used directly: Users can directly use native query statements from data sources in charts for data queries and display, without any conversion or adaptation.

3. Data security and privacy protection: To protect user data security and privacy, the platform does not store any added data source information. All data source configurations are saved directly in the user's local Func, ensuring data source information security and avoiding data leakage risks.

Monitoring¶

1. Alert Strategy > Configure Notification Rules by Member:

   • Supports configuring multiple group member notification rules to take effect in parallel;
   • Member notification rule configurations support defining effective time ranges. If multiple time ranges exist, they will be matched in order, and only the first matched time range's notification rules will be used for alert sending.

2. Monitor: Supports searching when configuring associated alert strategies.

3. Mute Time: When defining "Repeat" mute times, supports custom mute start time and mute duration, and supports configuring mute cycles "By Day," "By Week," and "By Month," providing more flexible mute time definitions. Also added a mute plan preview function to view currently defined mute times.

Billing¶

High Consumption Alert:

• Supports custom configuration of alert notification members. When billing items exceed the set threshold, email alerts will be sent to the Owner and the corresponding notification members;
• Supports setting exclusive notification members for each billing item;
• Supports creating external email addresses as notification members by pressing Enter.

Application Performance Monitoring¶

1. RUM > Analysis Dashboard, Container > Analysis Dashboard: View switching display optimization provides a smoother experience when users switch views.
2. Log built-in pages: Supports searching when selecting log indices, optimizing the operation experience.

Events¶

1. Unrecovered Event Explorer supports adjusting the event query range via the time widget;
2. Event Explorer supports custom configuration of display columns.

Scenarios¶

Chart optimization: Supports configuring whether to use scientific notation for chart data display.

Management¶

Blacklist: Added name and description fields to distinguish usage and other related scenarios.

Help Center¶

Help documentation optimization: Integration pages now include descriptive information to help intuitively view integration information.

Deployment Plan Updates¶

1. Supports modifying configuration files to customize the default time range for explorers;
2. Management backend member information supports entering international phone numbers.

New Integrations¶

• Added Azure MySQL;
• Added Huawei Cloud Mariadb integration;
• Added Huawei Cloud EIP integration;
• Added Huawei Cloud WAF integration;
• Added Confluent cloud integration;
• Updated Alibaba Cloud SAE integration, adding trace and log parts integration;
• Updated SQLSERVER monitor.

Bug Fixes¶

1. Fixed the issue where some explorer list column names could not be defined as aliases via the "Display Columns" entry;
2. Fixed the issue where applications listed during RUM metric generation were not affected by data access rules;
3. Fixed the front-end style adaptation issue in Application Performance Monitoring > Error Tracking > Issue Auto Discovery;
4. Fixed the issue with the Resource Catalog query time in Dashboard > Charts;
5. Fixed the issue where the billing analysis account period column width could not be adjusted.

November 20, 2024¶

Feature Updates¶

Microsoft Cloud Marketplace Listing¶

TrueWatch has been listed on the Overseas Microsoft Cloud Marketplace, and Commercial Plan workspaces now support "Microsoft Cloud" as a settlement method.

Integration > Extensions¶

• Func platform linkage optimization: All connected DataFlux Funcs can be viewed in the extensions, making it easier for users to manage and monitor all connected external data sources.

Deployment Plan Updates¶

• Volcengine TLS supports JSON indexing and search.

Bug Fixes¶

1. Fixed the issue where forwarding data to Kafka consumed excessive Kafka outbound traffic.

November 13, 2024¶

OpenAPI Updates¶

1. Snapshots: Supports creating snapshots for log explorers via OpenAPI Create Snapshots.
2. Data Access: OpenAPI supports obtaining the data access list, supports Get/Create/Modify individual data access rules.

Feature Updates¶

Billing¶

1. High Consumption Alert optimization: High consumption alert now includes total consumption alert functionality, supporting setting alert amounts for total workspace consumption.

Application Performance Monitoring¶

• Java applications support creating memory snapshots, helping developers quickly analyze and optimize application performance.
• Installation guide optimization: Added K8S deployment installation guide.

DQL Query¶

Supports directly specifying [today], [yesterday], [this week], [last week], [this month], [last month] time parameters.

Management¶

1. Invite members: Added line break separation, allowing multiple emails to be recognized via line breaks.
2. Data forwarding rule expansion: Data forwarding to Huawei Cloud OBS supports AK, SK access, providing richer OBS access methods.

3. Workspace:

   • Workspace deletion or disbanding now includes a 7-day temporary storage logic, with data being cleared after 7 days;
   • Supports configuring default and pinned spaces at the personal account level.

Monitoring¶

1. Monitor enhancement: When selecting logs for detection, supports searching log indices for selection.
2. Monitor: Monitor threshold trigger judgment configuration is appended to events for recording, with the associated event field: df_monitor_checker_value_with_unit, and the detection value with unit can be rendered in the event content via {{ Result_with_unit }}.
3. Intelligent monitoring events support bilingual switching: When switching workspace languages, intelligent monitoring events will change accordingly.
4. Cloud billing monitoring view: Cloud billing views support more instance-level billing analysis.

Scenarios¶

1. Bar charts now include Y-axis upper and lower limit settings for more precise chart display control;
2. Cross-workspace queries now support selecting "All Spaces" to retrieve all information at once.

Deployment Plan Updates¶

1. Forgot password: When forgetting the login password, supports retrieving the password via the account's associated email verification code. Note: If the account is not associated with an email, this method cannot be used.
2. Global DCA Configuration: Added global DCA address configuration, allowing one-click configuration of DCA addresses for all workspaces.

New Integrations¶

• Added Azure SQL Servers;
• Added Huawei Cloud RDS SQLServer;
• Added Huawei Cloud DDS integration;
• Huawei Cloud DCS (redis) added slow log collection process;
• Huawei Cloud MongoDB slow log collection process;
• Huawei Cloud RDS MYSQL slow log collection process;
• Optimized Node Exporter view, documentation;
• Optimized EMQX view, documentation, added monitor;
• Optimized Kubernetes view;
• Optimized SQLServer view and monitor;
• Optimized Redis view and monitor;
• Optimized Kafka view and monitor.

Bug Fixes¶

1. Fixed the issue where setting filter conditions in alert strategies and using extended fields in event information for filtering could not send alerts.
2. Fixed the issue where filter conditions in alert strategies did not display all content.
3. Fixed the issue where regular expression matching in alert strategies caused monitors to fail to send alerts.
4. Fixed the issue where configuring alert delivery to incident tracking and notifying specified Webhooks via incident tracking did not send issue.add type creation notifications.
5. Fixed the issue where secondary menu data in the Resource Catalog was unstable.
6. Fixed the issue where after adding a resource classification to the secondary menu in the Resource Catalog, the resource classification was deleted but the secondary menu remained.
7. Fixed the issue where JSON in the Resource Catalog was cleared after saving.
8. Fixed the issue where resources in the Resource Catalog incorrectly associated dashboards via tags.
9. Fixed the issue where loading display columns in the Resource Catalog was not smooth, with obvious transitions between display columns.
10. Fixed the issue where log flow chart filter conditions were lost after version upgrades, causing views to fail to load.
11. Fixed the issue where the "Add Filter" function in the log explorer could not fully recognize trace_id strings.
12. Fixed the issue where -bpf_net_l7_log logs were inaccurately associated with network logs.

November 6, 2024¶

Feature Updates¶

Price Adjustment¶

1. User Access PV 30-day, 60-day gradient price adjustment;
2. Application Performance Trace 30-day, 60-day gradient price adjustment;
3. Session Replay price adjustment.

Monitoring¶

Alert strategies now support defining notification rules based on member range, helping users better manage alert notifications and issue handling boundaries.

Scenarios¶

Rollup Function is only applicable to metric data queries. In chart simple mode, query selections for other data types will be deprecated.

Bug Fixes¶

1. Fixed the issue where the infrastructure honeycomb chart did not display specific usage rates.

October 30, 2024¶

New Integrations¶

• Alibaba Cloud SAE;
• Node Exporter;
• Azure Public IP;
• Grafana Dashboard;
• Greenplum.

Feature Updates¶

Cloud Billing¶

Added primary navigation menu, with pre-configured explorers and billing analysis views for cloud billing data, optimizing user viewing experience.

Note: Free Plan does not support this.

Monitoring¶

1. Host Intelligent Monitoring added network detection extension: Provides efficient network performance monitoring based on host network monitoring, helping users monitor host network traffic in real-time, identify abnormal traffic and potential connection issues, and provide early warnings to avoid affecting normal business operations. The system supports multi-dimensional visualization to help users deeply analyze and understand host network usage, optimize bandwidth allocation and resource utilization, and provide data support for future capacity planning, ensuring reasonable configuration of network resources.

2. Monitor enhancement:

3. Supports batch setting associated alert strategies in the monitor list.

4. Log explorers can one-click configure 【Log Detection】 type monitors based on current filter and search conditions.

5. Note: Monitors only support left* queries if both the site and workspace levels have left* queries enabled. Otherwise, if the log explorer is configured with left* queries, jumping to the monitor will result in query errors.

6. Notification Targets list:

7. Added search and quick filter functions to quickly retrieve notification targets;

8. Notification targets that fail to send for two consecutive days and are disabled by the system will display a mark after the name.

9. Mute Rules optimization:

10. Added rule name and description configuration functions to improve rule management convenience;

11. Event attributes support logical combination relationships (AND and OR) between different fields;

12. Optimized list display effects, supports custom display columns, improving user interface personalization experience.

13. Alert Strategies:

14. Label matching logic in notification rules supports free combination of AND and OR relationships between different fields, with the same interaction experience as the explorer filter search component;
15. Added custom operation permission configuration;
16. Added alert strategy description field.

Scenarios¶

1. Added Host NET analysis view: By aggregating host network usage, bandwidth allocation, and resource utilization metrics, it provides data support for future capacity planning, ensuring reasonable configuration of network resources.
2. Chart query optimization:
3. index does not support by group queries, optimizing DQL query interaction experience;
4. By tag range / filter tag range listing is precise to the metric level;
5. DQL query added function to obtain log indices: show_logging_index(), which can be applied in dashboard view variables, and chart query index configuration supports view variable filling;
6. Chart expression queries now support cross-workspace queries;
7. Optimized time lock display in composite charts, providing a more intuitive user experience.
8. Explorer page optimization: The bound host tab page in the explorer details page now displays host_ip.

Pipeline¶

Pipeline text processing optimization: Separated "Local Pipeline" and "Central Pipeline," allowing different types to add processing scripts for the same data source. Added prompt information to help users more intuitively understand processing differences.

Management¶

Data Access function integration:

• Application performance and metrics added data access function;
• Management added "Data Access" function module, integrating all data types, supporting users to quickly query and filter.

Synthetic Tests¶

Synthetic test task optimization: Websocket synthetic test content box input limit increased to 128k; Synthetic test task page overall optimization.

Deployment Plan Updates¶

1. Management backend added monitor menu: Lists all workspace monitors, supports searching, filtering monitors; supports modifying monitor enable/disable status, deleting, exporting monitors, and supports cloning single/batch monitors to selected workspaces.
2. Note: Composite detection monitors do not support cross-workspace cloning.
3. MFA security authentication optimization: Added hidden 7-day auto-login option switch, supports configuring whether to enable no-authentication login options.
4. CDN domain configuration: Can configure CDN domain in the configuration file, RUM application access page will automatically obtain and display.

Bug Fixes¶

1. Fixed the issue where pie charts displayed 0% values ambiguously, optimizing the proportion and visual effect of 0% values in drawing.
2. Fixed the issue where the default page loading error when viewing the "Events" menu.
3. Fixed the issue when selecting multiple log indices.
4. Fixed the issue where extra_data field was missing when passing Events from "External Event Monitors" to third-party tools.
5. Fixed the issue where switching hosts via quick filter in the log explorer did not refresh the right-side data.
6. Fixed the issue where table charts optimized the display of null values.
7. Fixed the issue where API-imported data access rule query conditions were not displayed on the page.
8. Fixed the issue where synthetic test logs could be viewed in the log function but could not find related data sources during PL processing.
9. Fixed the issue where application performance monitoring service map reported errors.
10. Fixed the issue where DQL did not support tag calculations after upgrading to the latest version of the self-developed time-series database.
11. Fixed the issue where logs could not be located when entering from the log context.
12. Adjusted application intelligent monitoring sensitivity, reducing excessive request count anomalies.
13. Improved mutation detection monitor's incompatibility with high-frequency tags.

October 16, 2024¶

New Integrations¶

• Hadoop hdfs datanode;
• Hadoop hdfs namenode;
• Hadoop yarn nodemanager;
• Hadoop yarn resourceManager;
• Fluent Bit;
• Azure VM;
• NPD: Added event mode collection & introduction.

Feature Updates¶

Obsy AI Agent¶

TrueWatch introduced Obsy AI Agent, providing quick responses and answering TrueWatch related questions.

Cloud Billing¶

1. Added cloud billing explorer: When billing data is collected into TrueWatch, cloud billing explorers can be created via templates to view all cloud billing data.
2. Added cloud billing system views: Through consumption analysis at the product, region, and instance levels, it helps users quickly analyze and understand current cloud resource consumption trends, providing reference for future cloud resource cost planning.
3. Added cloud billing intelligent monitoring: Cloud billing intelligent monitoring provides efficient cloud cost management tools, helping users monitor cloud service consumption in real-time, identify abnormal costs, and provide early warnings to avoid unnecessary expenses. It supports multi-dimensional visualization to help users analyze and understand cloud resource consumption patterns, providing data support for future budget planning, optimizing cloud resource allocation, and ensuring cost efficiency.

Monitoring¶

1. Monitor Configuration interaction optimization:
2. Supports configuring incident tracking Issue association creation for selected level events;
3. Optimized Crontab custom detection frequency and detection interval configuration interaction;
4. Added notification content customization.
5. Alert strategy configuration optimization:
6. Added regular expression matching for event filter conditions within rules;
7. Optimized filter condition configuration display.
8. Data sampling optimization: On the monitor configuration page and metric analysis page, when charts automatically trigger data sampling due to large data volumes, users can manually turn off data sampling.

Logs¶

1. Logs added error tracking: Supports error log tracking and analysis.

Scenarios¶

1. Dashboard visibility range optimization: Dashboard visibility range supports custom configuration of visible members on top of public access, and supports batch setting dashboard visibility ranges to improve dashboard management efficiency.
2. Template Variables adjustment: Added #{startTime}, #{endTime} two time variables, supporting obtaining the actual query time of the current chart, which can be applied in chart queries and chart links.
3. Log flow chart display column optimization: If the display column of the log flow chart is a @json format cut-out field, units can be set for this field to unify chart data viewing and analysis experience.
4. Platform chart color scheme upgrade: Platform chart color schemes have been upgraded for better visual effects.
5. Cross-workspace queries: All chart expression queries now support cross-workspace queries.

Management¶

1. Blacklist function enhancement: Supports selecting data sources for all data types via select all, single select, and multiple select configurations.
2. Self-built indices/sensitive data scanning/data forwarding: To optimize data processing and writing performance, function rule configuration filter conditions removed "match" and "not match" matching modes.

Events¶

1. Unrecovered Event Explorer optimization:
2. Optimized batch operation interaction, added one-click select current page and one-click select all options, supporting quick recovery of selected abnormal events;
3. Manual recovery generates recovery events, and OK event titles display optimization.

Infrastructure¶

1. Resource Catalog optimization: Resource Catalog supports custom resource classification icons, providing rich icon selections to enhance user experience.

Deployment Plan Updates¶

1. Dashboard view variable dropdown list values removed limit 50 restriction to meet different data query needs.
2. Management backend update: Management menu hidden items added "Help," "Avatar," "System Notifications," and "Shortcut" options to improve management efficiency.

Bug Fixes¶

1. Fixed the issue where integrations could not be searched in the English version.
2. Fixed the issue where RUM snapshot sharing could not be accessed.
3. Fixed the issue where composite monitoring A&&B did not take effect, A||B generated Result only had A's value, not B's value.
4. Fixed the issue where view settings master-slave failed to match null values.
5. Fixed the issue where leaderboard charts could only retain 2 decimal places, and smaller actual values were directly rounded off.
6. Fixed the issue where scheduled report dashboard emails were incomplete.
7. Fixed the issue where continuous trigger judgment times greater than 10 could not be saved.
8. Fixed the issue where APM and log data could not be aggregated after migrating to the base.
9. Fixed the issue where management backend only displayed local login when both local and LDAP login methods were enabled.
10. Fixed the issue where synthetic tests had no data.
11. Fixed the issue where log exports did not have the message content column.
12. Fixed the issue where log sorting was abnormal.
13. Fixed the issue where selecting multiple indices in the log explorer did not display context logs.
14. Fixed the issue where Grafana chart conversion tool conversion efficiency compatibility was low.
15. Fixed the issue where adding fields in synthetic test > quantity statistics configuration appeared outside the input box.
16. Fixed the issue where storage log monitors could not obtain detection dimensions.
17. Fixed the issue where incident tracking configuration Issue discovery did not generate Issues after abnormal events.
18. Fixed the issue where dashboard view variable dropdown lists did not support keyword queries.
19. Fixed the issue where text could not be written to Doris during troubleshooting.
20. Fixed the issue where event status counts changed within the same time period.
21. Adjusted role query logic, fixing the issue where roles exceeding 100 could not be searched.

September 25, 2024¶

Breaking Changes¶

Synthetic test tags: Field name changed from tags.info to df_label.

Feature Updates¶

Scenarios¶

1. Dashboards added historical version records: Can view version records saved within the last three months for this dashboard. After selecting a version, you can view the chart details of that version and mark the changes of the selected version in JSON format with different colors.
2. Supports restoring the current dashboard to the selected historical version;
3. Supports cloning a certain version of the dashboard.
4. Custom Explorer supports changing display column order: When editing custom explorer configurations, selected fields can be dragged to change the field display order.

Monitoring¶

Monitors added custom operation permissions: After configuring operation permissions, only authorized objects can enable/disable, edit, and delete this monitor.

• Note: If operation permissions are not configured, monitor permissions follow the default "Monitor Configuration Management" permissions.

User Access Monitoring¶

1. Session replay enhancement: Mobile session replay functionality launched, supporting mobile applications to replay user behavior trajectories and locate usage issues.
2. Data access enhancement: User access added data access function, supporting application-level data view range configuration for fine-grained management of member data query boundaries.

Management¶

1. Cross-space authorization optimization: Updated the definition of login space to "Current Space" to simplify user operations across different spaces.
2. Data Forwarding Rules support specifying storage directories: When the archive type is AWS S3, Huawei Cloud OBS, or Alibaba Cloud OSS, supports forwarding data to the corresponding directory in the storage bucket for better management of forwarded data.
3. Note: Please be cautious when changing storage paths, as there is a delay of about 5 minutes in updating configurations, and some data may still be forwarded to the original directory after the change.
4. Data forwarding explorer optimization:
5. Time widget adjustment: Adjusted to the universal time widget for explorers, supporting precise minute-level forwarded data.
6. SLS query logstore data storage location adjustment: "Data Forwarding" no longer retains SLS query logstore data, which can be queried in the "Log Explorer."

Synthetic Tests¶

If the tag format is key:value, the key value will be extracted to the first-level field in the results.

• Note: If the custom tag key value conflicts with other synthetic test result attribute values, it will be discarded.

Infrastructure¶

1. Resource Catalog function enhancement: Optimized the resource catalog architecture, supporting user-defined resource explorer templates, and providing list and honeycomb chart view modes to enhance resource data viewing and analysis experience.
2. Container interaction optimization: When switching views in container and K8s infrastructure explorers, filter and search conditions are retained and not reset.

Help Center¶

Integrated the help center page, allowing users to access all integration lists through a unified entry, and supports search and tag filtering to improve search efficiency.

Bug Fixes¶

1. Fixed the issue where PromQL query result rule mappings in views did not take effect.
2. Fixed the issue where host survival alert v2 monitors occasionally failed to recover.
3. Fixed the issue where dashboard scheduled reports failed to open when sent to email.
4. Fixed the issue where synthetic test tasks created via OpenAPI could not be modified.
5. Fixed the issue where PV statistics in the billing plan did not match the PV quantity in the fee details.
6. Fixed the issue where user information was not carried after adding user information to Webhook.
7. Fixed the issue where view variable values did not match the values in the dashboard.
8. Fixed the issue where homepage data display in shared workspaces.
9. Fixed the issue where template variable value dropdowns in data-authorized spaces did not match the actual data obtained.
10. Fixed the issue where time-series table charts often failed to load data in dashboards.
11. Optimized several page display issues.

Deployment Plan Updates¶

1. Datakit Management supports filtering online hosts: Supports filtering hosts that reported data in the last 10 minutes, 15 minutes, 30 minutes, 1 hour, and 3 hours.
2. Management backend:
3. When the space storage type is "Volcengine," supports configuring hot data retention duration.
4. New workspace copy and interaction optimization, default main engine only has one option, users do not need to specify additionally.
5. Platform color scheme enhancement: Supports defining navigation bar and navigation text display color configurations.
6. License usage and expiration reminders: Added global prompts for License即将到期、已过期和使用超量等状态；针对平台组件版本升级新增控制台全局提示。

September 4, 2024¶

Feature Updates¶

Logs¶

1. Log explorer added interaction: Long press Ctrl to perform "Add to Query," "Exclude from Query," and "Copy" operations on explorer list text segmentation, facilitating quick query addition of target data. Similarly, the log details page content area also supports this interaction.
2. Logs support binding Volcengine TLS external indices: Can bind external indices to directly view and analyze Volcengine log data on the TrueWatch platform.
3. Logs > Data Access optimization:
4. Added data access navigation page, with a help document jump link added to the upper right corner of the page;
5. Data access rules added 【Name】 field as a unique ID, added rule 【Description】 field, allowing custom names and descriptions to distinguish rule usage scenarios.
6. Log explorer filter optimization: Log explorer list 【Log Index】 filter supports search, allowing keyword search for indices and selection.
7. Log status customization: Supports custom log status values, and setting colors for each status value in the console to adapt to different log access scenarios, making log status display more personalized and intuitive.

Scenarios¶

1. Added new chart type heatmap: Users can intuitively view data distribution and trends through color depth, helping better understand data.
2. Chart link redirection: Links Function functions with dashboard data to achieve link redirection.
3. Alert statistics chart component upgrade: Changed listing logic, the new version of the alert statistics chart listing effect is consistent with the unrecovered event explorer, unifying user viewing experience.
4. Note: Cross-workspace listing does not support "Create Issue" and "Manual Recovery" operations.
5. Grouped table chart custom display columns: In grouped table charts, added custom display column function, supporting setting display or hide for returned columns, providing more flexible data display methods.
6. Chart cross-space authorization query interaction optimization: After enabling space authorization, authorized space lists can be directly selected above the query, making operations more intuitive and convenient.

Incident Tracking¶

Incident tracking Issue and monitor events: If the content contains @ member configuration, it will carry @member account attribute information when passed externally via Webhook.

Monitoring¶

1. External event detection: Supports custom event occurrence time and tag attribute addition. Note:
2. Field values uniformly use string type;
3. Field names cannot start with _ underscore and df_ prefix and cannot conflict with current event default fields;
4. Custom tag field names cannot conflict with dimension_tags defined field names.
5. Terminology adjustment: "No data" officially updated to "Data gap," ensuring unified page configuration viewing experience.

Pipeline¶

Separated synthetic test data and log data processing, and added "Synthetic Tests" data type. Avoids mutual interference during data processing, ensuring data processing accuracy and efficiency.

Note: Only central Pipeline supports "Synthetic Tests" data type, and synthetic test node DataKit version must be >= 1.28.0.

Others¶

1. Workspace list display optimization: For list operation bar, added creator avatar hover prompt for creator, creation time, updater, update time; overall optimization of operation item display.
2. List added audit event jump: List operation bar added audit event jump link, clicking can jump to view corresponding audit events.
3. Account management display adjustment: Single sign-on users do not display password items in the 【Account Management】 menu.

Bug Fixes¶

1. Fixed the issue where local Function functions as data sources did not link time range with time widget during queries.
2. Fixed the issue where PromQL data query result display order was incorrect.
3. Fixed the issue where monitor and chart query filter components were not unified.
4. Fixed the issue where clicking related events from the incident tracking entry would display in non-intelligent monitoring event lists causing no query results.
5. Fixed the issue where Lark incident tracking Webhook integration had no data and no callback.
6. Fixed the issue where ten-thousand system unit display numbers.
7. Fixed the issue where adding filter conditions in explorer analysis mode could not search fields.
8. Fixed the issue where creating Issues from monitor events had no source link.
9. Fixed the issue where log export to CSV failed.
10. Fixed the issue where high consumption alert email notifications were not received.
11. Fixed the issue where switching channels in incident tracking did not change the Issue list.
12. Fixed the issue where log indices a, b had the same field name (case-insensitive), causing errors when selecting indices a, b in the log explorer.

Deployment Plan Updates¶

1. Chart custom return quantity: Time-series charts, pie charts, table charts, leaderboards, rectangle tree charts, and maps support custom return data quantities, with no maximum limit, to meet different data display needs.
2. Management backend > Datakit management, supports exporting Datakit list.

August 21, 2024¶

Breaking Changes¶

• OpenAPI / Global API: 【Events】 Unrecovered event data source changed from UE to E.

New Features¶

• Management: Added Client Token unified management entry. When users access RUM applications via public network DataWay, they can replace the system-generated Token with custom-created Tokens.

Continuous Optimization¶

• Incident Tracking:
• Added Issue Discovery page. Through this feature, you can customize specific rules for Issue discovery, and uniformly manage and filter abnormal events and related data triggered by monitor detection rules. Treat a series of events as caused by a single reason, set filter conditions for these events, and select aggregation dimensions to further refine the data. After refinement, the data will be aggregated based on the detection frequency you set. Finally, the system will automatically push this information to the specified channel based on the preset Issue title and description, ensuring all relevant parties can receive and effectively handle these Issues.
• Configuration Management > Notification Strategies: Notification strategy lists added creation/update information display.

Regular Updates¶

• Data retention policy:
• Original 【Application Performance】 item split into 【Application Performance-Trace】, 【Application Performance-Profile】, supporting users to separately configure Trace data and Profile data retention policies;
• Original 【Data Forwarding】 name changed to 【Data Forwarding-TrueWatch】.
• Monitoring > Notification Target Management: Continuous one-day sending failure will trigger system notifications; continuous two-day sending failure will trigger system notifications and automatically disable.
• Unrecovered Event Explorer:
• Data source changed to query event data, using df_fault_id as the unique identifier for aggregation, and returning the latest data result for display.
• Page overall UI transformation.
• Application Performance Monitoring (APM) > Trace: Service Call Relationship Chart adjusted to service topology display, and displays the number of calls between services.
• Management:
• Added 【Workspace Description】;
• In edit mode, interaction changed to open a new window;
• Workspace list supports searching and locating by workspace name or description.
• Logs > BPF Logs > Seven-layer BPF Network Logs: Network request topology chart UI optimization, highlighting the distinction between server and client.
• Synthetic Tests > HTTP Tests > Advanced Settings > Request Settings default added Accept-Encoding:identity.

Deployment Plan Updates¶

• Added Synthetic Test Node Management entry, supports creating platform-level synthetic test nodes, and uniformly managing all nodes through the node list. Nodes created through this entry support configuring Chinese and English node names, adapting to TrueWatch domestic and international site displays and reported data result content.

• Data retention policy:

• Considering user storage cost factors, custom data retention duration is needed. Deployment plan management backend added 【Session Replay】 configuration item.

• Original 【Data Forwarding】 name changed to 【Data Forwarding-Default Storage】;

• Original 【Application Performance】 item split into 【Application Performance-Trace】, 【Application Performance-Profile】, supporting users to separately configure Trace data and Profile data retention policies;

• Supports Volcengine TLS as the underlying data storage engine.

BUG Fixes¶

• Fixed the issue where incident tracking notification strategies did not take effect;
• Fixed the issue where APM trace export was abnormal;
• Fixed the issue where modifying notification targets via OpenAPI reported no permission configuration display;
• Fixed the issue where log explorers could not automatically obtain source filter after resetting the time range;
• Fixed the issue where source filter range was already added in the explorer search bar, but "Quick Filter" still displayed all source outside the filter conditions;
• Fixed the issue where mutation detection reported errors;
• Fixed the issue where after writing data access rules via OpenAPI, the UI page could not view role information;
• Fixed the issue where chart data format settings did not take effect in chart legends;
• Fixed the issue where associated synthetic test tasks were deleted under self-built synthetic test nodes, but actual synthetic tests were still running.

August 7, 2024¶

New Features¶

• User Access Monitoring (RUM): Added heatmap. Visually presents visitor interactions with the website, obtaining click data and click positions of page elements to understand user focus points.

• Application Performance Monitoring (APM) > Trace: Added installation guide page.

• Monitoring: Added detection type——interval detection V2, predicting normal fluctuation ranges based on historical data confidence intervals of detection metrics.

Continuous Optimization¶

• Incident Tracking:
• Configuration Management > Notification Strategies: Added operation audit and execution log viewing entry. When receiving Issue notifications, sometimes notifications are not sent normally or there are doubts about notification strategies, you can view the current notification strategy operation audit events and execution log data to judge.
• APM / RUM Issue Auto Discovery supports adding filter conditions;
• For deployment plans, added unified management of incident tracking levels entry;
• Calendar:
  • In the calendar edit page, different notification targets automatically generate colors;
  • Calendar Management: 【My Calendar】 and 【All Calendars】 added statistical quantities;

Regular Updates¶

• Monitoring > SLO:
• Added tag configuration, ultimately acting on generated event data information;
• Configuration optimization: By setting 【Target】 and 【Minimum Target】 7-day compliance rate ranges, judge generating warning or urgent events;
• Supports sending alert notifications by associating 【Alert Strategies】.
• Synthetic Tests:
• Synthetic test tasks added tag configuration;
• Configure synthetic test task page test module optimization;
• List added quick filter module;
• Explorer: Analysis mode supports exporting CSV files.
• Infrastructure > Network: Network detail page > Network analysis supports ip:port and ip two dimension statistics listing display.
• Application Performance Monitoring > Service > Create/Modify Service List: Added format validation when filling in repository links and document links.

Breaking Changes¶

• Management > Attribute Claims: Custom attribute field values adjusted to string type for storage.

Deployment Plan Updates¶

• Template Management: Added import template entry, template scope includes view templates, monitor templates, custom explorer templates, Pipeline. Supports deployment plan users changing custom templates to platform-level templates for other workspaces to use.
• User Management: Batch operation interaction optimization.

August 15, 2024¶

TrueWatch Updates¶

• Monitoring
• Monitor > Mutation Detection > Detection Metrics: Supports 【Compared to Previous Period】 option, enabling final comparison of data within a fixed time period.
• Mute Management: Added 【Additional Information】 function, supporting adding explanations to mute rules to identify the reason or source of muting.
• Intelligent Monitoring > Host Intelligent Monitoring: Added network traffic and disk IO two detection dimensions.
• Scenarios > Dashboard:
• View Variables: Edit page style optimization, supports defining dropdown single select and multiple select.
• Grouped Table Charts, Metric Analysis > Table Charts support multi-column query result display adaptation, such as

L::RE(`.*`):(count(*),message,host) {index = 'default' and status = 'error'} BY source,service

• Explorers:
• Log Explorer > Context Logs supports microsecond-level data query filtering, solving the issue where multiple data entries at the same moment (millisecond) could not locate and display a specific log context.
• All explorers support selecting export data quantity as CSV files.
• Added explorer search query audit events, i.e., user-initiated query operations will be recorded in audit events.
• Service Management: Migrated from the original path 【Scenarios】 to 【Application Performance Monitoring】, optimizing user experience.
• Generate Metrics: Supports configuring multiple by groups, with no quantity limit.
• DQL Query: Expression queries support specifying value filling, supports filling sub-query results and final values.
• User Access Monitoring > Android: Application configuration display optimization.
• Events: Added detail page associated view jump entry. In the absence of detection dimension data, can jump to explorer view in the detail page.

TrueWatch Deployment Plan Updates¶

• Added DataKit List Management page.
• Deployment plan configuration single sign-on connection supports custom login display title, description, and logo.
• Users: Added extended attribute configuration.
• Supports local users directly configuring attributes in the edit page.
• Supports single sign-on automatically appending third-party user attributes to TrueWatch via userinfo interface by default.

April 24, 2024¶

TrueWatch Updates¶

• Management:
• Added Cloud Account Management: Centralizes all enterprise cloud service accounts for unified management, and distinguishes by the uniqueness of certain configurations under the account. By configuring integrated collectors, independently manages cloud services under each account, achieving fine-grained control of business data.
• Account Management: Account Login Expiration Time adjustment.
• Added Quick Search popup, can quickly view the most recently accessed pages within the current workspace and other function-related pages.
• Infrastructure > Containers: Added Statefulset and Persistent Volumes two object explorers.
• Incident Tracking:
• Added Issue Responsible Person configuration, TrueWatch will send email notifications to responsible persons.
• Channel Management: Supports upgrade configuration. That is, set when new Issues exceed a specific number of minutes, if no responsible person is specified, send upgrade notifications to corresponding notification targets.
• Monitoring
• Monitor: Detection Configuration: Supports configuring in 【Trigger Conditions】 that after triggering conditions take effect multiple times in a row, trigger generating events again.
• Mute Management:
  • Mute rule list page display optimization: Supports listing all mute rules in the current workspace, can quickly filter and list target rules via quick filters.
  • Event attribute matching supports reverse selection, filter condition formats as follows: attribute:value, attribute:*value*, -attribute:value, -attribute:*value*, different field combination relationships are AND, multiple values of the same field are OR.
• DQL match function meaning changed to exact match. This change only applies to the new engine, respectively applied to explorer and monitor scenarios.
• Explorer scenario example: host:~cn_hangzhou.
• Monitor scenario example:

  window("M::`cpu`:(avg(`load5s`)) { `host` = match('cn-hangzhou.172.16.***') } BY `host`", '1m')
  

• Scenarios > Dashboard Charts can be directly downloaded as PNG images, table charts can also be exported as CSV files.
• Logs > Bind Index: 【Field Mapping】 changed to non-mandatory.
• Integration/Built-in Views: Template added tag management.
• Service Map cross-workspace node style display adjustment.

TrueWatch Deployment Plan Updates¶

• Management > Basic Information > License Information: DataKit quantity limit supports adjustment based on data statistics range, changed to hosts or DKs with survival time >= 12 hours.
• Supports configuring blacklists, custom selecting import TrueWatch integration, view templates, monitor template ranges.

OpenAPI Updates¶

• Pipelines Add/Modify: Added profiling type;
• User Views Add/Modify: Supports binding dashboard configuration.

April 10, 2024¶

TrueWatch Updates¶

• Monitoring > Monitor > Create: Added 【Data Gap】, 【Information Generation】 configuration areas to better distinguish abnormal data and data gap situations.
• Management: Added System Notifications page, can view all configured abnormal status messages under the current account's workspaces.
• Scenarios:
• Chart Queries: Added Rollup Function, this function also applies to 【Metric Analysis】 and 【Query Tools】;
• Dashboard/User Views: Added pin pin function. When the current access workspace is authorized to view data from several other workspaces, supports setting queries of other workspace data as default options.
• System Views: Supports cloning creation as dashboards or user views;
• Custom Explorers: Optimized search mode; in non-edit mode, hover over 【Data Range】 to view all filter conditions.
• Explorer > Quick Filter:
• Added 【Dimension Analysis】 button, clicking can quickly switch to explorer analysis mode;
• Supports adding current field to display columns/removing from display columns by clicking external buttons.
• Free Plan Workspace > Navigation Bar: Added 【Upgrade Now】 button.
• Infrastructure > Containers > Honeycomb Chart: Added CPU Usage (Standardized) and MEM Usage (Standardized) two metric fill methods.

TrueWatch Deployment Plan Updates¶

Workspace Management: Added Data Reporting Limit, helping stakeholders save resource usage costs.

March 27, 2024¶

TrueWatch Updates¶

• Monitoring:
• Alert Strategy Management: Each notification rule (including default notifications and custom notifications) configuration added supports upgrade notification conditions.
• Monitor > Event Content: Added Custom Advanced Configuration, supports adding associated logs and error stacks;
• Host Intelligent Monitoring: Changed current mutation display to anomaly alerts based on periodic predictions, trend charts display current metrics and confidence interval upper and lower bounds, anomalies exceeding the confidence interval are marked in red.
• Scenarios > Charts: Added Topology Chart.
• APM > Trace Detail Page > Service Call Relationship: Adjusted to service topology display, and displays the number of calls between services.
• Data Retention Policy: Session Replay data retention policy is linked with RUM retention policy, i.e., RUM data retained for 3 days, Session Replay data also retained for 3 days.
• Explorers:
• Event Explorer > Basic Attributes: Added detection metric display configuration, cached locally, globally adapted;
• APM > Error Tracking > Clustering Analysis > Detail Page: Supports creating incident tracking Issues;
• RUM > Error > Clustering Analysis > Detail Page: Supports creating incident tracking Issues;
• RUM > View > Detail Page:
  • Performance: Added 【All Entries】 option, lists all associated data under the current View;
  • Fetch/XHR: Clicking data rows supports opening corresponding trace detail pages or Resource detail pages.
• Time Widget: When obtaining "current time," precise to milliseconds.
• Management > Ticket Management:
• Added star rating and evaluation;
• Feedback tickets need to be adjusted to automatically close if no customer feedback within 7 days;
• Supports ticket list export;
• Tickets with status 【Completed】, 【Revoked】 can be restarted;
• Tickets submitted under accounts after account cancellation that are in unclosed status support automatic closure processing.
• Cloud market opening process overall optimization.

TrueWatch Deployment Plan Updates¶

• Data Retention Policy: Supports workspace owners configuring data retention policies, and supports custom input retention duration. Application scenarios:
• Metric Management > Measurement;
• Logs > Index > Create.
• Users: Supports inviting members via email accounts.

March 13, 2024¶

TrueWatch Updates¶

• Monitoring > Monitor: Monitor type 【Composite Detection】 launched. Supports combining multiple monitor results into one monitor through expressions, and ultimately alerts based on the combined result.
• Scenarios > Service: Supports cross-workspace Service Map queries.

TrueWatch Deployment Plan Updates¶

• Management > Basic Information: Added "Used DK Quantity" display;
• Management > Users: Added Grouping page, based on groups can configure associated workspaces and roles, users can obtain access permissions to corresponding workspaces through groups.

March 6, 2024¶

TrueWatch Updates¶

• Monitoring
• Monitor > Detection Frequency: Added Crontab Custom Input, meeting the need to execute detection only during specific time periods;
• Mutation Detection: Added 【Last 1 Minute】, 【Last 5 Minutes】 detection intervals;
• Mute Management: When selecting mute range, "Event Attributes" is non-mandatory, can configure more granular matching rules as needed.
• DataFlux Func: Added Function External Function. Allows third-party users to fully utilize Function's local cache and local file management service interfaces to write functions, executing data analysis queries within the workspace.
• APM > Trace:
• Title area UI display optimization;
• For flame charts, waterfall charts, Span lists exceeding 10,000 Span results, supports viewing unshown Spans via offset settings;
• Added Error Span filter entry; supports inputting Span corresponding resource names or Span IDs for search matching.
• Scenarios
• Charts: Added Sankey Chart;
• View Variables: Added Select button, checked by default selects all current values, can deselect as needed.
• Account Management: Added Cancellation entry.
• Explorers:
• UI display optimization;
• Filters added regular expression matching / reverse regular expression matching mode;
• Wildcard filters and searches support left * matching.
• Events > Detail Page: 【Alert Notifications】 tab page UI display optimization.

TrueWatch Deployment Plan Updates¶

• Added Login Method Selection, unified management of login methods;
• Added Delete operation for local accounts and single sign-on accounts.

January 31, 2024¶

TrueWatch Updates¶

• Monitoring:
• Intelligent Monitoring:
  • Host, Log, Application Intelligent Detection frequency adjusted to execute every 10 minutes, each execution calculated as 10 call fees;
  • To improve algorithm accuracy, Log, Application Intelligent Detection uses data offloading, enabling one intelligent monitor generates corresponding measurement and metric data. This adjustment generates additional time series, specifically the number of current monitor configuration filter conditions filtered detection dimensions (service, source) * detection metric quantity, as monitor filter conditions are not stored, if monitor filter condition configurations are modified, new equivalent time series are generated, so there will be time series duplicate billing on the day of modification, normal resumes the next day.
• Alert Strategy Management:
  • Added custom notification time configuration, refining alert notification configurations by cycle, time range;
  • Repeated alerts added 【Permanent】 event option.
• Monitor
  • Alert Configuration: Supports configuring multiple alert strategies; if multiple are configured, df_monitor_name and df_monitor_id will appear in multiple forms, separated by ;;
  • Incident Tracking Issue linkage transformation: Added 【Event Recovery Synchronously Closes Issue】 switch, when abnormal events recover, incident tracking Issues are synchronously recovered;
  • Monitor list added clone button.
• Notification Target Management: Added Simple HTTP Notification Type, directly receiving alert notifications via Webhook address;
• Scenarios:
• Charts: Unit added 【Currency】 option; Advanced Configuration > Period Comparison changed to 【Period Comparison】;
• Service Management > Resource Calls: Leaderboard added TOP / Bottom quantity selection.
• Explorers: Display column settings added 【Time Column】 switch.
• Billing:
• Workspace lock popup page added Create Workspace entry, optimizing operation experience;
• AWS registration process optimization.

TrueWatch Deployment Plan Updates¶

• Supports LDAP Single Sign-On;
• Workspace Management > Data Storage Policy added custom options, range <= 1800 days (5 years); among them, metrics added optional 720 days, 1080 days retention durations; in console settings > edit data storage policy, modify and save to synchronously update backend data storage;
• Users: Supports one-click configuration of distributing workspaces and roles to user accounts;
• Added console audit event viewing entry, can quickly view all workspace-related operation audits;
• Added 【Management Backend MFA Authentication】.

January 11, 2024¶

TrueWatch Updates¶

• Logs:
• Added BPF network log collection and log detail page, supports JSON format conversion; detail page added readable display mode;
• Added binding 【Associated Network Logs】;
• Data Access: Added batch operations.
• Scheduled Reports: Added optional sharing methods 【Public Share】 or 【Encrypted Share】.
• Dashboard:
• View Variables added 【All Variable Values】 parameter passing option;
• Time-Series Charts: Added sorting logic (only new engine), supports sorting returned results.
• Generate Metrics: Supports batch operations; members with standard and above permissions support cloning.
• Monitor:
• Notification Target Management: Adapted new DingTalk robot, when creating, "Key" option is non-mandatory, quickly associate DingTalk robot.
• SLO deduction logic optimization.
• User Access Monitoring (RUM): Public network Dataway supports IP conversion to geographic location information.## 2023 年 12 月 25 日

OpenAPI 更新¶

1. 资源目录：支持获取资源目录列表，支持创建/修改/删除/导出资源查看器。
2. 生成指标：支持列出指标列表，支持新建/修改/启用/禁用/删除指标。

功能更新¶

Pipeline¶

自动编写 Pipeline：传统的日志解析方式往往依赖于对规则的熟悉进行手动编写脚本，这不仅消耗大量时间，也容易出错，导致编写 Pipeline 的效率低下，无法及时响应和解决问题。为了提高效率并降低人工成本，本功能利用大模型的自然语言处理能力，辅助自动化解析日志内容，帮助开发人员在编写 Pipeline 时更加高效，精准地从大量日志中提取有价值的信息，简化数据流和处理过程。

监控¶

AI 告警压缩合并：通过大模型对监控告警事件进行压缩合并，减少告警事件的冗余，特别是当系统在短时间内产生大量告警时。通过对告警事件的整合，确保运维人员在指定的时间周期内（如 10 分钟）仅接收到一条综合的告警信息，而非多条重复的告警信息。

云账单分析¶

云账单场景视图、查看器新增消费货币类别展示。

场景¶

1. 仪表板、内置视图新增标识 ID：仪表板、用户视图可通过标识 ID 确定唯一，图表内配置跳转链接时可通过标识 ID 定位仪表板或者用户视图。
2. 查看器自动刷新逻辑调整：在时间组件中去除“暂停”按钮，刷新频率选项里新增 “off”，用户可按需关闭自动刷新功能，操作更简洁直观。同时取消个人设置中的“查看器自动刷新”开关，避免功能冗余，优化用户设置体验。
3. 查看器新增函数：MD5()，支持在查看器中查询脱敏方式为 MD5 加密的日志数据。
4. 图表优化：图表内支持获取 PromQL 指标名称做图例，可使用 {__name__} 配置指标名称做别名展示。
5. PromQL 编辑器新增语法高亮以及语法提示。

指标¶

生成指标优化：

1. 将所有类型的生成指标整合至指标目录下，方便用户统一查看与管理。
2. 支持对已有指标数据进行再聚合操作，以此生成全新指标，为数据分析提供更多维度。

管理¶

数据访问和敏感数据查看脱敏规则应用优化：优化多规则多角色组合叠加数据查看范围，若成员拥有多个角色最终数据访问的范围取规则内定义的范围合集，敏感数据查看脱敏规则同理。

基础设施¶

资源目录跨工作空间查询：在资源目录菜单中，实现支持跨工作空间查询资源列表数据。

部署版更新¶

1. 部署版支持自定义导航栏和操作按钮显示文案。

2. 管理后台成员菜单优化：

   • 新增“登录类型”列，支持区分默认登录成员和 SSO 登录成员，并支持以登录类型进行筛选；
   • 针对 SSO 登录成员，支持编辑“其他属性”字段，可自定义属性信息。此类信息在发送告警时，会同步到 Webhook 对外以及对应的事件内容区域。

3. 主存储引擎为火山引擎的工作空间，链路、日志和自定义日志索引的存储策略调整为：标准存储、低频存储、归档存储三项，可分别配置热数据、低频数据、归档数据存储时长。同时，若主存储引擎为火山引擎，在TrueWatch控制台新增索引时，存储策略同步调整为：标准存储、低频存储、归档存储。

新增集成¶

• 新增 火山引擎 ALB 集成；
• 新增 火山引擎 CLB 集成；
• 新增 火山引擎 NAS 集成；
• 新增 Kube State Metrics 集成；
• 新增 Ranger admin 集成；
• 新增 Ranger tagsync 集成；
• 新增 Ranger usersync 集成；
• 更新 Redis 仪表板和监控器；
• 更新 nginx 集成&仪表板；
• 更新 RabbitMQ 仪表板和监控器。

Bug 修复¶

1. 解决了主机标签不生效的问题；
2. 解决了控制台更新 token 失败的问题；
3. 解决了组合看板中排序因刷新操作而丢失的问题；
4. 解决了共享看板中表格图在编辑模式下无数据显示的问题；
5. 解决了在跨空间授权查询看板中组合图表切换不同图表没有发出后端请求的问题；
6. 解决了 RUM 视图中 version 数据异常的问题；
7. 解决了监控器内置 Func 提示找不到账号并报 404 错误且未产生事件的问题；
8. 解决了用户访问监测列表数据展示为 0 的问题；
9. 解决了 AI 智能助理页面下方 UI 显示的问题；
10. 解决了资源目录中的某个资源分类删除成功后依然存在的问题。
11. 解决了数据断档误告警的问题。

2023 年 12 月 11 日¶

功能更新¶

场景¶

1. 拓扑图新增外部数据查询：允许用户通过 DataFlux Func 实现外部数据绘制拓扑图。用户只需按照图表结构接入数据，即可轻松实现外部数据的可视化展示。
2. 图表优化
   • 图表显示效果优化：对图表的显示效果进行优化，调整为侧滑列出，分类展示，使图表的查找和使用更加便捷。
   • 图表描述及适用场景显示：图表列表中增加了图表描述及适用场景的显示，可以帮助用户更好地理解和选择合适的图表类型。
3. 视图变量优化
   • 视图变量支持配置值列出上限，避免数据列出过多导致页面加载性能问题。
   • 部署版默认列出上限为 50，支持自定义默认列出数量，注意：页面配置限制优先全局限制。

基础设施¶

1. 资源目录优化：

   • 新增分组功能：为了提高资源管理的效率，新增资源分组功能。用户可以将具有共性的资源分类进行分组，便于管理和查看，从而优化资源的组织结构。
   • 查看器蜂窝图模式优化：支持配置“颜色填充”和“分组分析”的字段列表，用户可以根据需要自定义可选的字段。

2. 容器、资源目录查看器搜索优化：容器查看器新增 container_name 搜索，资源目录新增 name 搜索。

监控¶

告警策略配置优化

• 支持通过告警策略入口一键创建关联监控器，新增以告警策略为中心的统一告警通知管理模式。
• 优化告警策略列表关联显示交互。

RUM¶

RUM 新增了对 React Native 应用类型的支持，并允许上传 SourceMap 以进行数据解析和还原。如果您需要为原本在 Android / iOS 类型下创建的 React Native 应用上传 SourceMap，请注意：

1. 在创建新应用时，选择 “React Native” 应用类型。在创建中，您可直接复制 Android / iOS 类型下已有的 React Native 应用的名称和应用 ID，然后点击“创建”按钮。完成创建后，您便可以在该应用下配置 SourceMap。这一变更仅涉及 React Native 应用的创建方式，不会对数据采集产生影响。

2. 如果您原先在 Android / iOS 类型下的 React Native 应用有关联的监控器，并且尚未在 “React Native” 类型下重新创建相同 ID 的 React Native 应用，原有的监控器可以继续正常工作。但如果您已经重新创建了 “React Native” 类型的应用，原有的监控器在“应用名称”项将无法获取数据，显示为空。在这种情况下，您需要选择应用类型为 “React Native” 并重新保存此监控器。

APM¶

APM 安装引导新增自动注入方式：在 APM（应用性能监测）的安装引导中，新增了 Kubernetes Operator 自动注入的安装方式。这种方式简化了 APM 的部署流程，使得用户可以更快捷地在 Kubernetes 环境中安装和使用。

新增集成¶

• 新增 HBASE region 集成；
• 新增 HBASE master 集成；
• 优化 NodeExporter 集成视图；
• 新增 华为云 DCAAS 云专线 集成；
• K8S dashboard 调整；
• 更新 memcached 视图和监控器；
• 更新 rabbitmq 视图和监控器。

Bug 修复¶

1. 解决了表达式查询数值异常的问题；
2. 解决了在图表用 PromQL 查询某一段时刻启动的 pod 的 cpu 使用率时，用时序图在时间范围内可以看到数据，但在转换为图表查询时数据查询不出来的问题。
3. 解决了告警事件不恢复的问题。
4. 解决了 PromQL 查询结果异常的问题。
5. 解决了私有化部署 > 管理后台修改热存储时长时报错 warmretention 字段缺少的问题。
6. 解决了部署版日志搜索结果和火山引擎 TLS 上搜索不一致的问题。
7. 解决了仪表板大屏在嵌入用户的 IFrame 页面时会出现随机缩放的问题。
8. 解决了用户管理新建分组将其加入多个空间后页面卡顿的问题。
9. 解决了首次进入TrueWatch控制台 > 事件时报错 df_fault_id 的问题。
10. 解决了打开事件菜单时会提示超出内存限制的问题。
11. 解决了概览图表达式计算错误的问题。
12. 解决了监控器触发告警后未产生事件的问题。
13. 解决了部署版火山引擎底座日志查询功能异常的问题。

2023 年 11 月 27 日¶

Breaking Changes¶

1. OpenAPI：若通过 API 配置告警策略按成员配置通知规则模式，需注意通过 OpenAPI 方式新增/修改成员类型的告警策略的参数结构调整。

2. 仪表板 > 可见范围：新增“自定义”选项，支持配置此仪表板的操作、查看权限成员。

   • 注意：若您先前在可见范围处添加了“团队”，团队配置将失效，需重新配置。

功能更新¶

云账单¶

1. 新增功能引导页：提供简洁明了的步骤及说明，让用户能够迅速上手；
2. 新增支持火山引擎、微软云数据接入。

外部数据源接入¶

1. MySQL 数据存储系统支持：平台现已支持接入 MySQL 数据存储系统，用户可以利用此功能实现数据的实时查询和分析。

2. 原生查询语句直接使用：用户可以直接在图表中使用数据源的原生查询语句进行数据查询和展示，无需进行任何转换或适配。

3. 数据安全与隐私保护：为了保护用户的数据安全和隐私，平台不会存储任何添加的数据源信息。所有数据源配置将直接保存在用户的本地 Func 中，确保数据源信息的安全，避免数据泄漏风险。

监控¶

1. 告警策略 > 按成员配置通知规则：

   • 支持配置多组成员通知规则并行生效；
   • 成员配置通知规则支持定义生效的时间范围，若存在多组时间范围则按照序号顺序匹配，多组时间范围最终只会取第一个匹配到的时间范围内通知规则做告警发送。

2. 监控器：配置关联告警策略时支持搜索。

3. 静默时间：定义“重复”静默时间时，支持自定义静默开始时间、静默持续时长，支持配置“按天”、“按周”、“按月”的静默周期，帮助更灵活地定义静默时间。同时新增静默计划预览功能，可以查看当前定义的静默时间。

付费计划与账单¶

高消费预警：

• 支持自定义配置预警通知成员，当计费项超出设定阈值时，会向 Owner 和对应的通知成员发送邮件预警；
• 支持在每一个计费项下设置此计费项的专属通知成员；
• 支持回车创建外部邮箱作为通知成员。

应用性能监测¶

1. RUM > 分析看板、容器 > 分析看板：视图切换显示优化，用户在切换视图时能获得更流畅的体验。
2. 日志内置页面：在选择日志索引时支持搜索，优化操作体验。

事件¶

1. 未恢复事件查看器支持通过时间控件调整查询事件范围；
2. 事件查看器支持自定义配置显示列。

场景¶

图表优化：支持配置图表数据展示是否使用科学计数法进位。

管理¶

黑名单：新增名称和描述项，支持区分用途和其它关联场景。

帮助中心¶

帮助文档优化：集成页面新增描述信息，帮助直观查看集成信息。

部署版更新¶

1. 支持修改配置文件以自定义查看器默认时间范围；
2. 管理后台成员信息支持输入国际手机号码。

新增集成¶

• 新增 Azure MySQL；
• 新增 华为云 Mariadb 集成；
• 新增 华为云 EIP 集成；
• 新增 华为云 WAF 集成；
• 新增 Confluent cloud 集成；
• 更新 阿里云 SAE 集成，添加 链路、日志部分的集成；
• 更新 SQLSERVER 监控器。

Bug 修复¶

1. 修复部分查看器列表列名无法通过 “显示列” 入口进行别名定义的问题；
2. 修复了 RUM 生成指标时，所列出应用未受数据访问规则影响的问题；
3. 修复了应用性能监控 > 错误追踪 > Issue 自动发现前端样式适配的问题；
4. 修复了仪表板 > 图表内资源目录查询时间的问题；
5. 修复了账单分析账期列宽度无法调整的问题。

2023 年 11 月 20 日¶

功能更新¶

微软云市场上架¶

TrueWatch已在海外微软云市场上架，同时商业版工作空间新增“微软云”结算方式。

集成 > 扩展¶

• Func 平台联动优化：可以在扩展中查看所有接入的 DataFlux Func 列表，方便用户管理和监控所有已接入的外部数据源。

部署版更新¶

• 火山云 TLS 支持 JSON 索引和搜索。

Bug 修复¶

1. 解决了数据转发到 Kafka，消耗了过多 Kafka 出网流量的问题。

2023 年 11 月 13 日¶

OpenAPI 更新¶

1. 快照：支持日志查看器以 OpenAPI 方式创建快照。
2. 数据访问：OpenAPI 支持获取数据访问列表，支持获取/新建/修改单个数据访问规则。

功能更新¶

付费计划与账单¶

1. 高消费预警优化：高消费预警新增总消费预警功能，支持针对工作空间总消费设置预警金额。

应用性能监测¶

• Java 应用支持创建内存快照，帮助开发者可以快速分析和优化应用性能。
• 安装引导优化：新增 K8S 部署安装引导。

DQL 查询¶

支持直接指定 [today]、[yesterday]、[this week]、[last week]、[this month]、[last month] 时间参数。

管理¶

1. 邀请成员：新增换行分隔功能，多个邮箱之间可通过换行识别。
2. 数据转发规则拓展：数据转发到华为云 OBS 支持 AK、SK 方式访问，提供更丰富的 OBS 访问方式。

3. 工作空间：

   • 工作空间删除或解散新增 7 天暂存逻辑，7 天后工作空间内数据再进行最终清除；
   • 支持配置个人账号级别的默认空间和置顶空间。

监控¶

1. 监控器功能增强：检测指标选择日志时，支持搜索日志索引进行选取。
2. 监控器：监控器阈值触发判断配置追加到事件中做记录，关联事件字段：df_monitor_checker_value_with_unit，事件内容内可通过 {{ Result_with_unit }} 渲染附带单位的检测值。
3. 智能监控事件支持双语切换：当切换工作空间语言的时，智能监控事件支持随着工作空间语言更改。
4. 云账单监控视图：云账单视图支持更多实例视角的账单分析。

场景¶

1. 柱状图新增 Y 轴上下限设置，可以更精确地控制图表的显示效果；
2. 在进行跨工作空间查询时，支持选择“全部空间”，以便用户一次性获取全部信息。

部署版更新¶

1. 忘记密码：当忘记登录密码时，支持以账号关联的邮箱验证码的方式找回密码。注意：若账号未关联邮箱，则无法通过此方式找回密码。
2. 全局 DCA 配置：新增全局 DCA 地址配置，可一键配置所有工作空间的 DCA 地址。

新增集成¶

• 新增 Azure SQL Servers；
• 新增 华为云 RDS SQLServer；
• 新增 华为云 DDS 集成；
• 华为云 DCS（redis） 新增慢日志采集流程；
• 华为云 MongoDB 慢日志采集流程；
• 华为云 RDS MYSQL 慢日志采集流程；
• 优化 Node Exporter 视图、文档；
• 优化 EMQX 视图、文档，添加监控器；
• 优化 Kubernetes 视图；
• 优化 SQLServer 视图和监控器；
• 优化 Redis 视图和监控器；
• 优化 Kafka 视图和监控器。

Bug 修复¶

1. 解决了告警策略设置过滤条件，使用事件信息中的扩展字段进行过滤，无法发送告警的问题。
2. 解决了告警策略里面的过滤条件未展示所有内容的问题。
3. 解决了告警策略正则匹配导致监控器无法发出告警的问题。
4. 解决了配置告警投递到异常追踪，并通过异常追踪通知到指定 Webhook，但检查通知结果发现 issue.add 类型的创建通知未能正常发送的问题。
5. 解决了资源目录的二级菜单数据不稳定的问题。
6. 解决了资源目录 > 添加资源分类到二级菜单后，资源分类删除而二级菜单还在的问题。
7. 解决了资源目录的 JSON 在保存后，下次打开会清空掉配置的问题。
8. 解决了资源目录中，通过标签为资源关联仪表板，未命中标签的资源错误关联上该仪表板的问题。
9. 解决了选择资源目录，加载显示列不连贯，中间有明显的过渡显示列的问题。
10. 解决了版本升级后，日志流图中筛选条件丢失，视图无法正常加载的问题。
11. 解决了日志查看器 "添加筛选" 功能无法完整识别 trace_id 字串的问题。
12. 解决了 -bpf_net_l7_log 日志的关联网络日志不准确的问题。

2023 年 11 月 6 日¶

功能更新¶

价格调整¶

1. 用户访问 PV 30 天、60 天梯度价格调整；
2. 应用性能 Trace 30 天、60 天梯度价格调整；
3. 会话重放价格调整。

监控¶

告警策略新增根据成员范围定义通知规则，帮助用户更好的管理告警通知和问题处理边界。

场景¶

Rollup 函数仅适用于指标数据查询，在图表简单模式下，对其他数据类型的查询选择将做下线处理。

Bug 修复¶

1. 解决了基础设施蜂窝图不显示具体的使用率的问题。

2023 年 10 月 30 日¶

新增集成¶

• 阿里云 SAE；
• Node Exporter；
• Azure Public IP；
• Grafana Dashboard；
• Greenplum。

功能更新¶

云账单¶

新增一级导航菜单，针对云账单数据预置查看器和账单分析视图，优化用户查看体验。

注意：体验版暂不支持。

监控¶

1. 主机智能监控新增网络检测扩展：基于主机的网络监控提供了高效的网络性能监测，帮助用户实时监控主机的网络流量，识别异常流量和潜在的连接问题并及时预警，避免影响业务正常运行。系统支持多维度可视化功能，帮助用户深入分析和理解主机的网络使用情况，优化带宽分配和资源利用率，为未来的容量规划提供数据支持，从而确保网络资源的合理配置。

2. 监控器功能增强：

3. 支持在监控器列表中批量设置关联告警策略。

4. 日志查看器可针对当前筛选和搜索条件一键配置【日志检测】类型监控器。

5. 注意：只有在站点和工作空间级别都开启了 左* 查询的前提下，监控器才支持 左* 查询。否则日志查看器若配置 左* 查询，跳转到监控器会查询报错。

6. 通知对象列表：

7. 新增搜索、快捷筛选功能，支持快速检索通知对象；

8. 针对连续两天发送失败被系统禁用的通知对象，名称后展示标记。

9. 静默规则优化：

10. 新增规则名称和描述配置功能，提升规则管理的便捷性；

11. 事件属性支持不同字段的逻辑组合关系（AND 和 OR）；

12. 优化列表显示效果，支持自定义显示列，提升用户界面的个性化体验。

13. 告警策略：

14. 通知规则内标签匹配逻辑支持不同字段自由组合 AND 和 OR 的关系，交互体验同查看器筛选搜索组件一致；
15. 新增自定义操作权限配置；
16. 新增告警策略描述填写。

场景¶

1. 新增主机 NET 分析视图：通过对主机的网络使用情况，带宽分配和资源利用率等指标的汇聚，为未来的容量规划提供数据支持，从而确保网络资源的合理配置。
2. 图表查询优化：
3. index 不支持做 by 分组查询，优化 DQL 查询交互体验；
4. By 标签范围 / 筛选标签范围列出精确到指标级别；
5. DQL 查询新增获取日志索引的查询函数：show_logging_index()，可在仪表板视图变量处应用，同时图表查询索引配置支持视图变量填充；
6. 图表的表达式查询功能现已支持跨空间查询；
7. 优化组合图表的时间锁定显示，提供更加直观的用户体验。
8. 查看器页面优化：查看器详情页中绑定主机的 Tab 页追加 host_ip 显示。

Pipeline¶

Pipeline 文本处理优化：隔离“本地 Pipeline” 和“中心 Pipeline”，允许不同类型添加同一个数据源的处理脚本。并新增提示信息，帮助用户更直观地了解处理差异。

管理¶

数据访问功能整合：

• 应用性能和指标新增数据访问功能；
• 管理中新增「数据访问」功能模块，整合所有数据类型，支持用户快速查询与过滤。

可用性监测¶

拨测任务优化：Websocket 拨测内容框输入限制提升到 128k；拨测任务页面整体优化。

部署版更新¶

1. 管理后台新增监控器菜单：列出所有工作空间的监控器，支持搜索、筛选监控器；支持修改监控器的启用/禁用状态、删除、导出监控器等操作，同时支持克隆单个/批量监控器到选中工作空间。
2. 注意：组合检测监控器不支持跨工作空间克隆。
3. MFA 安全认证优化：新增隐藏 7 天自动登录选项的开关，支持配置免认证登陆选项是否开启。
4. CDN 域名配置： 可以在配置文件中配置 CDN 域名，RUM 应用接入页面将自动获取并显示。

Bug 修复¶

1. 解决了饼图的 0% 值显示歧义的问题，已优化 0% 值在画图时的占比和视觉效果。
2. 解决了查看“事件”菜单的默认页面加载报错的问题。
3. 解决了日志索引选择多索引时出现的问题。
4. 解决了通过“外部事件监控器”传入第三方工具的 Event 时 extra_data 字段缺失的问题。
5. 解决了日志查看器中通过快捷筛选方式切换主机，右侧数据不刷新的问题。
6. 解决了表格图中空值显示优化的需求。
7. 解决了 API 导入数据访问规则的查询条件在页面上不显示的问题。
8. 解决了拨测日志在日志功能中可以查看到，但在进行 PL 处理时找不到相关数据源的问题。
9. 解决了应用性能监测服务拓扑报错的问题。
10. 解决了自研时序数据库升级到最新版本后 DQL 不支持 tag 计算的问题。
11. 解决了通过日志上下文点进来无法定位到当前日志的问题。
12. 调整应用智能监控灵敏度，减少请求数异常突降过多问题。
13. 改善突变检测监控器对高频 tags 的不适配问题。

2023 年 10 月 16 日¶

新增集成¶

• Hadoop hdfs datanode；
• Hadoop hdfs namenode；
• Hadoop yarn nodemanager；
• Hadoop yarn resourceManager；
• Fluent Bit；
• Azure VM；
• NPD：新增事件模式采集&介绍。

功能更新¶

Obsy AI 智能体¶

TrueWatch引入 Obsy AI 智能体，提供快速响应，解答TrueWatch相关的问题。

云账单¶

1. 新增云账单查看器：当账单数据收集到TrueWatch后，可以通过云账单查看器模版创建云账单查看器查看到云账单的全部数据。
2. 新增云账单系统视图：通过对云账单的产品、地域、实例级别的消费分析，帮助用户快速分析和了解当前云资源的消费趋势，为未来云资源费用规划提供参考。
3. 新增云账单智能监控：云账单智能监控提供高效的云成本管理工具，帮助用户实时监控云服务消费，识别异常费用并预警，避免不必要支出。它支持多维度可视化功能，帮助用户分析和理解云资源的消费模式，为未来预算规划提供依据，从而优化云资源配置，确保费用物尽其用。

监控¶

1. 监控器配置交互优化：
2. 支持针对选中等级事件配置异常追踪 Issue 关联创建；
3. 优化 Crontab 自定义检测频率和检测区间配置交互；
4. 新增通知内容自定义。
5. 告警策略配置优化：
6. 规则内针对事件过滤条件新增正则匹配；
7. 优化过滤条件配置显示。
8. 数据采样优化：在监控器配置页面和指标分析页面，当图表因数据量过大自动触发数据采样时，用户可以手动关闭数据采样功能。

日志¶

1. 日志新增错误追踪：支持错误日志追踪查看分析。

场景¶

1. 仪表板可见范围优化：仪表板可见范围在公开的基础上，支持自定义配置可见成员，同时支持批量设置仪表板可见范围，以提高仪表板管理效率。
2. 模板变量调整：新增 #{startTime}、#{endTime} 两个时间变量，支持获取到当前图表实际查询时间，可在图表查询、图表链接中应用此变量。
3. 日志流图显示列优化：日志流图的显示列若为 @json 格式切出字段，可对此字段设置单位，以统一图表数据的查看和分析体验。
4. 平台图表配色升级：平台图表配色得到升级，提供更佳的视觉效果。
5. 跨工作空间查询：所有图表内的表达式查询功能现已支持跨工作空间查询。

管理¶

1. 黑名单功能增强：支持对所有数据类型的数据来源进行全选、单选、多选配置。
2. 自建索引/敏感数据扫描/数据转发：为优化数据处理和写入性能，涉及功能规则配置过滤条件去掉 ”match“ 和 ”not match“ 匹配模式。

事件¶

1. 未恢复事件查看器优化：
2. 优化批量操作交互，新增一键勾选当前页和一键勾选全部选项，支持快速恢复当前选中的异常事件；
3. 手动恢复产生恢复后，OK 事件标题显示优化。

基础设施¶

1. 资源目录优化：资源目录支持自定义资源分类图标，提供丰富的图标选择，以提升用户体验。

部署版更新¶

1. 仪表板视图变量下拉框列出值取消 limit 50 的限制，以满足不同的数据查询需求。
2. 管理后台更新：管理菜单隐藏项新增“帮助”、“头像”、“系统通知”和“快捷入口”选项，以提高管理效率。

Bug 修复¶

1. 解决了英文版集成无法搜索的问题。
2. 解决了 RUM 快照分享无法访问的问题。
3. 解决了组合监测 A&&B 不生效，A||B 生成的 Result 只有 A 的值，没有 B 的值的问题。
4. 解决了视图设置主从时匹配空值失败的问题。
5. 解决了排行榜图表展示的数据中只能保留 2 位小数，实际数值较小的数据直接被四舍五入丢弃的问题。
6. 解决了定时报告仪表板邮件显示不全的问题。
7. 解决了开启连续触发判断次数大于 10 次时无法保存的问题。
8. 解决了 APM 和日志迁入底座后数据无法聚合的问题。
9. 解决了管理后台同时开启本地和 LDAP 登录方式，前台只显示本地登录的问题。
10. 解决了可用性监测无数据的问题。
11. 解决了日志导出没有 message 内容这一列的问题。
12. 解决了日志排序异常的问题。
13. 解决了日志查看器中选择多索引就不显示上下文日志的问题。
14. 解决了 Grafana 图表转换工具转化效率兼容性低的问题。
15. 解决了可用性检测 > 数量统计配置时，添加的字段在输入框外的问题。
16. 解决了存储日志监控器获取不到检测维度的问题。
17. 解决了异常追踪配置 Issue 发现后在产生异常事件后没有产生 Issue 的问题。
18. 解决了仪表板视图变量下拉列表不支持关键字查询的问题。
19. 解决了文本无法正常写入 Doris 排查的问题。
20. 解决了同一时间段事件状态数量会变的问题。
21. 调整角色查询逻辑，解决了角色数量超出 100 时搜索不到的问题。

2023 年 9 月 25 日¶

Breaking Changes¶

拨测标签：字段名由 tags.info 调整为 df_label。

功能更新¶

场景¶

1. 仪表板新增历史版本记录：可查看此仪表板三个月内保存的版本记录，选中某版本后，可查看此版本图表详情，并以 JSON 格式、通过不同颜色标记选中版本的变更内容。
2. 支持将当前仪表板还原至选中的历史版本；
3. 支持克隆仪表板某个版本。
4. 自定义查看器支持变更显示列顺序：自定义查看器编辑配置时，可拖拽所选字段变更字段展示顺序。

监控¶

监控器新增自定义操作权限：配置操作权限后，仅被赋予权限的对象可对此监控器进行启用/禁用、编辑、删除操作。

• 注意：若未配置操作权限，则监控器权限跟随“监控器配置管理”默认权限。

用户访问监测¶

1. 会话重放增强：移动端上线会话重放功能，支持移动端应用通过会话重放的方式回溯用户行为轨迹、定位使用问题。
2. 数据访问增强：用户访问新增数据访问功能，支持针对应用级别配置数据查看范围，精细化管理成员数据查询边界。

管理¶

1. 跨空间授权优化：将登录空间的定义更新为“当前空间”，以简化用户在不同空间间的操作。
2. 数据转发规则支持指定存储目录：存档类型为 AWS S3、华为云 OBS、阿里云 OSS 时，支持将数据转发到存储桶下的对应目录中，帮助更好地管理转发数据。
3. 注意：请谨慎变更存储路径，由于更新配置存在 5 分钟左右的延迟，变更后可能会有部分数据依然转发到原目录下。
4. 数据转发查看器优化：
5. 时间控件调整：调整为查看器通用时间控件，可获取精确到分钟的转发数据。
6. SLS query logstore 数据存储位置调整：“数据转发”不再保留 SLS query logstore 的数据，可在“日志查看器”中查询。

可用性监测¶

标签格式若为 key:value，在结果中会将 key 值提取到一级字段。

• 注意：若自定义的标签 key 值与其他拨测结果属性值重复，则做丢弃操作。

基础设施¶

1. 资源目录功能增强：优化了资源目录架构，支持用户自定义资源查看器模板，并提供列表和蜂窝图两种视图模式，以提升资源数据的查看和分析体验。
2. 容器交互优化：容器及 K8s 基础设施查看器切换查看时，筛选、搜索条件保留不重置。

帮助中心¶

整合了帮助中心页面，允许用户通过统一入口访问所有集成列表，并支持搜索及标签过滤功能，以提高查找效率。

Bug 修复¶

1. 解决了视图中 PromQL 查询结果设置规则映射不生效的问题。
2. 解决了主机存活告警 v2 监控器偶发无法恢复的问题。
3. 解决了仪表板定时报告发送到邮件时，报告中仪表板打开失败的问题。
4. 解决了使用 OpenAPI 创建的拨测任务无法修改的问题。
5. 解决了付费计划中 PV 统计量和费用详情里的 PV 数量不一致的问题。
6. 解决了 Webhook 添加用户信息后未携带手机号码的问题。
7. 解决了视图变量的取值和仪表板中的取值不一致的问题。
8. 解决了共享工作空间的首页数据显示问题。
9. 解决了数据授权空间中，模板变量值下拉框获取的数据与实际获取的数据对不上的问题。
10. 解决了时序表格图在仪表板中时常加载不出来数据的问题。
11. 优化了若干页面显示问题。

部署版更新¶

1. Datakit 管理支持筛选在线主机：支持筛选最近 10 分钟、最近 15 分钟、最近 30 分钟、最近 1 小时、最近 3 小时内有数据上报的主机。
2. 管理后台：
3. 空间存储类型为“火山引擎”时，支持配置热数据保留时长。
4. 新建工作空间文案和交互优化，默认主引擎只有一个选项时用户无需额外指定。
5. 平台配色模板增强：支持定义导航栏及导航文字显示颜色配置。
6. License 使用和过期提醒：针对 License 即将到期、已过期和使用超量等状态新增控制台全局提示；针对平台组件版本升级新增控制台全局提示。

2023 年 9 月 4 日¶

功能更新¶

日志¶

1. 日志查看器新增交互：长按 Ctrl 可针对查看器列表文本分词进行 “添加到查询”、“从查询中排除”、“复制” 操作，便捷目标数据的快速查询追加。同理日志详情页内容区域也支持此交互。
2. 日志支持绑定火山引擎 TLS 外部索引：可绑定外部索引，在TrueWatch平台直接查看及分析火山引擎日志数据。
3. 日志 > 数据访问优化：
4. 新增数据访问导航页，页面右上角新增帮助文档跳转链接；
5. 数据访问规则新增【名称】字段为唯一性 ID，新增规则【描述】字段，可自定义名称及描述以区分规则使用场景。
6. 日志查看器筛选项优化：日志查看器列表【日志索引】筛选项支持搜索，可关键字搜索索引并进行勾选。
7. 日志状态自定义：支持自定义日志状态值，并在控制台中为每个状态值设置颜色，适应不同的日志接入场景，同时使日志的状态显示更加个性化和直观。

场景¶

1. 新增图表类型热力图：用户可以通过颜色深浅直观地查看数据的分布和趋势，帮助更好地理解数据。
2. 图表链接重定向跳转：联动 Function 函数结合仪表板数据实现链接重定向跳转。
3. 告警统计图表组件升级：更换列出逻辑，新版本的告警统计图列出效果同未恢复事件查看器一致，统一用户查看体验。
4. 注意：跨工作空间列出情况下不支持 “创建 Issue” 和 “手动恢复” 操作。
5. 分组表格图自定义显示列：在分组表格图中，新增自定义显示列功能，支持对返回的列设置显示或隐藏，提供更灵活的数据展示方式。
6. 图表跨空间授权查询交互优化：开启空间授权功能后，可以直接在查询上方选择被授权的空间列表，操作更加直观和便捷。

异常追踪¶

异常追踪 Issue 和监控器事件：内容若存在 @ 成员配置，则通过 Webhook 对外传递时 Issue 或事件会同步携带 @成员的账号属性信息。

监控¶

1. 外部事件检测：支持自定义事件发生时间和标签属性追加。注意：
2. 字段值统一使用字符串类型；
3. 字段名称不支持以 _ 下划线和 df_ 前缀开头且不能与当前事件默认字段重名；
4. 自定义标签字段名不能与 dimension_tags 内定义字段名重名。
5. 名词调整：“无数据” 正式更新为 “数据断档”，保证页面配置查看体验统一。

Pipeline¶

将拨测数据和日志数据分离处理，并新增“可用性拨测”数据类型。避免在数据处理时的相互干扰，确保数据处理的准确性和效率。

注意：仅中心 Pipeline 支持“可用性拨测”数据类型，且拨测节点的 DataKit 版本必须 >= 1.28.0。

其他¶

1. 工作空间列表显示优化：针对列表操作栏，新增显示创建人头像鼠标悬浮提示创建人、创建时间、更新人、更新时间；同时对操作项整体显示进行优化。
2. 列表新增跳转审计事件：列表操作栏新增审计事件跳转链接，点击可跳转查看对应审计事件。
3. 账号管理显示调整：单点登录用户【账号管理】菜单下不显示密码项。

Bug 修复¶

1. 解决了本地 Function 函数做为数据源 查询时时间范围未联动时间控件问题。
2. 解决了 PromQL 数据查询结果 展示顺序不正确的问题。
3. 解决了监控器和图表查询 的筛选组件不统一的问题。
4. 解决了从异常追踪入口点击智能监控的相关事件会在非智能监控事件列表中显示导致无查询结果的问题。
5. 解决了飞书异常追踪 Webhook 集成无数据、无回调的问题。
6. 解决了万进制单位显示数字的问题。
7. 解决了查看器分析模式下，添加筛选条件时无法搜索字段的问题。
8. 解决了从监控事件创建 Issue 无来源链接的问题。
9. 解决了日志导出到 CSV 失败的问题。
10. 解决了未收到高消费预警邮件通知的问题。
11. 解决了异常追踪切换频道，Issue 的列表无变化的问题。
12. 解决了日志索引 a、b 存在相同字段名(忽略大小写)时，日志查看器中勾选索引 a、b 出现报错 的问题。

部署版更新¶

1. 图表自定义返回数量：时序图、饼图、表格图、排行榜、矩形树图和地图支持自定义返回数据的数量，无最大限制，以满足不同的数据展示需求。
2. 管理后台 > Datakit 管理，支持导出 Datakit 清单。

2023 年 8 月 21 日¶

Breaking Changes¶

• OpenAPI / 全局 API：【事件】未恢复事件数据源从 UE 变更为 E。

新增功能¶

• 管理：新增 Client Token 统一管理入口，用户使用公网 DataWay 接入 RUM 应用时，可更换系统默认生成的 Token，使用自定义创建的 Token。

持续优化¶

• 异常追踪：
• 新增 Issue 发现页面。通过这一功能，您可以定制 Issue 发现的具体规则，对监控器检测规则触发的异常事件和相关数据进行统一管理和筛选。将一系列事件视为由单一原因引起，并为这些事件设置筛选条件，然后选择聚合维度来进一步细化数据。细化后，数据会根据您设定的检测频率进行聚合。最终，系统会根据您预设的 Issue 标题和描述，自动将这些信息推送到指定的频道，确保所有相关方都能及时接收并有效处理这些 Issue。
• 配置管理 > 通知策略：通知策略列表新增创建/更新的信息显示。

常规更新¶

• 数据保存策略：
• 原【应用性能】项拆分为【应用性能-链路】、【应用性能-Profile】，支持用户分别配置 Trace 数据和 Profile 数据的保存策略；
• 原【数据转发】名称修改为【数据转发-TrueWatch】。
• 监控 > 通知对象管理：连续一天发送失败会发系统通知；连续两天发送失败会发系统通知且自动禁用。
• 未恢复事件查看器：
• 数据源变更为查询事件数据，以 df_fault_id 作为唯一标识进行聚合，获取最近一条数据结果返回展示。
• 页面整体 UI 改造。
• 应用性能监测（APM） > 链路：服务调用关系图新增绑定内置视图能力，点击服务的卡片，即可快速查看与该服务关联的相关用户视图。
• 管理：
• 新增【工作空间描述】；
• 编辑模式下，交互变更为打开新窗口；
• 工作空间列表下支持通过工作空间的名称或描述来搜索定位。
• 日志 > BPF 日志 > 七层 BPF 网络日志：网络请求拓扑图 UI 优化，突出了服务端与客户端的区分。
• 可用性监测 > HTTP 监测 > 高级设置 > 请求设置默认添加 Accept-Encoding:identity。

部署版更新¶

• 新增拨测节点管理入口，支持创建平台级别拨测节点，并通过节点列表统一管理所有节点。通过此入口创建的拨测节点支持配置中英文节点名，从而适配TrueWatch的国内外站点显示和上报数据结果内容。

• 数据保存策略：

• 考虑到用户处于存储成本等因素的考量，需要自定义这些数据的保存时长，部署版管理后台新增【会话重放】配置项。

• 原【数据转发】名称修改为【数据转发-默认存储】；

• 原【应用性能】项拆分为【应用性能-链路】、【应用性能-Profile】，支持用户分别配置 Trace 数据和 Profile 数据的保存策略；

• 支持火山引擎 TLS 做为底层数据存储引擎。

BUG 修复¶

• 解决异常追踪的通知策略未生效的问题；
• 解决应用性能监测链路追踪导出异常的问题；
• 解决通过 OpenAPI 修改通知对象报错无权限配置显示的问题；
• 解决日志查看器重新设置时间范围后不能自动获取 source 筛选的问题；
• 解决查看器搜索栏已添加 source 筛选条件范围，但在“快捷筛选”中依旧显示过滤条件外的全部 source 的问题；
• 解决突变检测报错的问题；
• 解决通过 OpenAPI 写入数据访问规则后，UI 页面打开无法查看角色信息的问题；
• 解决图表设置的数据格式对图例中数据不生效的问题；
• 解决自建拨测节点下，关联的拨测任务删除后，实际拨测还在运行的问题。

2023 年 8 月 7 日¶

新增功能¶

• 用户访问监测（RUM）：新增热图。以视觉方式呈现访客与网站的互动情况，获取页面元素的点击数据和点击位置，了解用户的关注点。

• 应用性能监测（APM） > 链路：新增安装引导页面。

• 监控：新增检测类型——区间检测 V2，以检测指标的的历史数据建立的置信区间来预测正常波动范围。

持续优化¶

• 异常追踪:
• 配置管理 > 通知策略：新增操作审计和执行日志查看入口。在接收 Issue 通知时，有时会遇到通知未正常发送或针对通知策略有疑议，此时可查看当前通知策略的操作审计事件和执行日志数据来进行判断。
• APM / RUM Issue 自动发现支持添加筛选条件；
• 针对部署版，新增统一管理异常追踪等级入口；
• 日程：
  • 在日程编辑页面，不同的通知对象会自动生成颜色；
  • 日程管理：【我的日程】与【所有日程】新增统计数量；

常规更新¶

• 监控 > SLO：
• 新增标签配置，最终作用到产生的事件数据信息内；
• 配置优化：通过设置【目标】和【最低目标】的 7 天达标率范围，判断生成警告或紧急事件；
• 支持通过关联【告警策略】实现告警通知发送。
• 可用性监测：
• 拨测任务新增标签配置；
• 配置拨测任务页面的测试模块优化；
• 列表新增快捷筛选模块；
• 查看器：分析模式下支持导出 CSV 文件。
• 基础设施 > 容器：新增进程关联页面展示。

BUG 修复¶

• 解决【任务调用】计费统计次数未显示的问题；
• 解决图表查询时【左 * 匹配】问题；
• 解决 BPF 网络日志返回数据未包含容器等相关信息的问题；
• 解决中心 Pipeline 失效问题。

Breaking Changes¶

• OpenAPI：
• SLO 创建/修改接口新增 tags、alertPolicyUUIDs 并弃用 alertOpt 参数；
• SLO 获取详情和列表接口返回结果中新增 tagInfo、alertPolicyInfos 字段，丢弃了 alertOpt 字段。

2023 年 7 月 24 日¶

TrueWatch更新¶

• 异常追踪：
• 新增分析看板：可视化展示不同指标数据。
• 新增日程管理和通知策略：对 Issue 的内容范围做进一步通知分配。
• 场景：
• 图表：新增时间偏移设置。启用时间偏移后，当查询相对时间区间时，实际查询时间范围向前偏移 1 分钟，以防止入库延迟导致数据获取为空。
• 仪表板：新增历史快照入口。
• 快照：保存快照时自动获取当前页面上选取的时间范围；分享快照时，可选择允许查看者更改时间范围。
• 监控：
• 基础设施存活检测 V2：新增附加信息。选定字段后，系统会做额外查询，但不会用于触发条件的判断。
• 通知对象管理：新增【操作权限】选项配置，由开关控制通知对象的操作（编辑、删除）权限。

2023 年 7 月 10 日¶

TrueWatch更新¶

• 场景：
• 视图变量：
  • 新增配置项开关：包含 * 选项。
  • 选择隐藏视图变量时，列表新增隐藏标识。
• 仪表板：分组组件支持配置颜色。
• 图表：别名功能覆盖排行榜、矩形树图、桑基图。
• 仪表板/查看器/内置视图：新增卡片属性信息，优化编辑配置。
• 监控：
• 通知对象管理：配置 Webhook 通知对象，支持追加成员信息。
• 应用智能检测：新增追踪历史变化，过滤周期性的异常数据突变；新增异常服务关联影响的用户数。
• 事件 > 事件列表查看器：显示列新增告警通知状态标识。
• 日志：
• 索引：绑定 SLS 外部索引时，新增访问类型选择，支持自定义公网访问或者内网访问；
• 日志查看器：聚类分析模式下，支持导出显示列数据及关联日志页面数据。

TrueWatch部署版更新¶

• 新增全局功能菜单管理配置，支持自定义控制台显示菜单范围，并同步至工作空间功能菜单栏。
• 模版管理：自定义模板管理上传逻辑优化。

2023 年 6 月 26 日¶

TrueWatch更新¶

• Pipelines：支持选择中心 Pipeline 执行脚本。
• 付费计划与账单：新增中心 Pipeline 计费项，统计所有命中中心 Pipeline 处理的原始日志的数据大小。
• 监控
• 通知对象管理：新增权限控制。配置操作权限后，仅被赋予权限的对象可对此通知对象进行编辑、删除操作。
• 智能监控 > 日志智能检测：新增追踪历史变化，过滤周期性的异常数据突变。
• 日志
• 数据访问：新增对被授权查看的日志索引做访问权限配置。
• 日志查看器：显示列拓展，支持添加 json 对象内字段内容到一级返回显示。
• BPF 网络日志：
  • 连接展示效果优化；
  • 支持直接跳转至详情页；
  • 支持自定义添加显示列。
• 场景
• 时序图：折线图、面积图新增断点连接设置，柱状图新增【显示返回值】按钮。
• 可用性监测：任务列表新增表头排序。
• DataFlux Func：支持TrueWatch异常追踪脚本集成钉钉应用。

TrueWatch部署版更新¶

Profile：通过配置参数，支持文件存储和对象存储两种方式。

2023 年 6 月 13 日¶

TrueWatch更新¶

• BPF 网络日志：优化 BPF 网络功能，增强 L4/L7 网络联动。
• APM/RUM：新增 【Issue 自动发现】功能。启用该配置后，TrueWatch会将符合配置项规则的错误数据记录自动创建 Issue。
• 监控
• 智能监控：新增 Kubernetes 智能检测：通过智能算法自动检测 Kubernetes 中的异常，检测指标包含 Pod 总数，Pod 重启，APIServer QPS 等。
• 告警策略管理：
  • 新增过滤功能。在进行告警规则配置时，该功能允许在原有等级基础上增加更细致的过滤条件，仅匹配等级+过滤条件的事件才会发送给对应的通知对象。
  • 支持选择外部邮箱做为通知对象。
• 监控器 > 事件内容：支持自定义输入外部邮箱。
• 场景
• 拓扑图：新增链接配置。
• 桑基图：由原先支持最多 4 个节点配置改为 6 个。
• Pipeline：列表新增过滤条件显示。
• 日志 > 索引：列表显示优化。

TrueWatch部署版更新¶

Profile：文件大小由原先的固定 5MB 修改为支持自定义，点击查看如何配置。

2023 年 6 月 3 日¶

TrueWatch更新¶

• 管理 > 跨工作空间授权：添加页面新增【数据范围】，支持多选数据类型。
• 日志 > 日志查看器：支持跨工作空间索引查询，快速获取其它空间的日志数据，从而突破日志数据存储位置的限制，大幅度提升数据分析和故障定位的效率。

2023 年 5 月 29 日¶

TrueWatch更新¶

• DCA
• 支持私有化部署，可直接通过工作空间页面按钮前往 DCA 控制台。
• 支持批量管理功能。
• 异常追踪：
• Webhook 接收通道：支持 Issue 回复的新增、修改通知；
• 支持选择团队或添加外部邮箱为 Issue 负责人。
• 日志 > 上下文日志：查询逻辑修改；支持通过日志上下文详情页对相关数据作进一步查询管理。
• 场景
• 视图变量：分组标题/图表标题支持配置 #{视图变量} 显示。
• 时序图 > 分析模式：支持调整时间间隔 interval。
• 仪表板/自定义查看器：当标签数量超过 20 个时，显示搜索框；标签名前显示颜色。
• 监控
• 监控器/智能监控/静默管理 > 快捷筛选：当值超过 20 个时，出现搜索框，支持搜索定位。
• 监控器：针对 PromQL 查询检测，事件通知中自动列出可使用的模板变量。
• 基础设施 > 网络：网络详情页 > 网络分析支持 ip:端口 和 ip 两种维度统计列出展示。
• 应用性能监测 > 服务 > 新建/修改服务清单：填写仓库链接、文档链接时增加格式校验。

Breaking Changes¶

• 管理 > 属性声明：自定义属性字段值调整为字符串类型进行存储。

TrueWatch部署版更新¶

• 模版管理：新增导入模版入口，模版范围包括视图模板、监控器模板、自定义查看器模板、Pipeline。支持部署版用户将自定义模板变更为平台级别模板供其他工作空间使用。
• 用户管理：批量操作交互优化。

2023 年 5 月 15 日¶

TrueWatch更新¶

• 监控
• 监控器 > 突变检测 > 检测指标：支持【环比上期】选项，从而实现某个固定时间段内的数据进行最终比较。
• 静默管理：新增【附加信息】功能，支持针对静默规则添加解释说明，从而标识静默的原因或者来源等信息。
• 智能监控 > 主机智能监控：新增网络流量、磁盘 IO 两项检测维度。
• 场景 > 仪表板：
• 视图变量：编辑页面样式优化，支持定义下拉单选、多选。
• 分组表格图、指标分析 > 表格图支持多列查询结果显示适配，如

L::RE(`.*`):(count(*),message,host) {index = 'default' and status = 'error'} BY source,service

• 查看器：
• 日志查看器 > 上下文日志支持微秒级的数据查询过滤，解决出现同一时刻（毫秒）有多条数据，导致不能命中定位显示某条日志上下文的问题。
• 所有查看器支持选择导出数据量为 CSV 文件。
• 新增查看器搜索查询审计事件，即由用户手动发起的查询操作会计入审计事件记录。
• 服务管理：由原来所属的路径【场景】迁移至【应用性能监测】，优化使用体验。
• 生成指标：支持配置多个 by 分组，不做数量限制。
• DQL 查询：表达式查询支持指定值填充，支持针对子查询做结果填充和最终值填充。
• 用户访问监测 > Android：应用配置显示优化。
• 事件：新增详情页关联查看跳转入口。在不存在检测维度数据的情况下，可在详情页跳转查看器查看。

TrueWatch部署版更新¶

• 新增 DataKit 清单管理页面。
• 部署版配置单点登录对接时，支持自定义登录显示标题、描述和 logo。
• 用户：新增扩展属性配置。
• 支持本地用户直接在编辑页面配置属性。
• 支持单点登录时默认自动将第三方用户属性配置通过 userinfo 接口追加到TrueWatch。

2023 年 4 月 24 日¶

TrueWatch更新¶

• 管理：
• 新增云账号管理：将企业所有的云服务账号集中起来进行统一管理，并借由账号下某些配置的唯一性来进行区分。通过配置集成采集器，针对每个账号下的云服务进行独立管理，从而实现对业务数据的精细化控制。
• 账号管理：账号登录过期时间调整。
• 新增快速搜索弹窗，可快速查看当前工作空间内最近访问的页面和其他各功能相关页面。
• 基础设施 > 容器：新增 Statefulset、Persistent Volumes 两种对象查看器。
• 异常追踪：
• 新增 Issue 负责人配置，TrueWatch会为负责人发送邮件通知。
• 频道管理：支持升级配置。即，设置新 Issue 超过某特定分钟数时，若未指定负责人，则发送升级通知给对应的通知对象。
• 监控
• 监控器：检测配置：支持在【触发条件】配置连续多次判断触发条件生效后，再次触发生成事件。
• 静默管理：
  • 静默规则列表页展示优化：支持列出当前工作空间所有静默规则，可通过快捷筛选快速过滤列出目标规则。
  • 事件属性匹配支持反选，筛选条件格式如下：attribute:value、attribute:*value*、-attribute:value、-attribute:*value*，不同字段组合关系为 AND，相同字段的多个值为 OR。
• DQL match 函数的含义变更为完全匹配。此变更仅针对新引擎，分别应用查看器、监控器这两个场景。
• 查看器场景示例：host:~cn_hangzhou。
• 监控器场景示例：

  window("M::`cpu`:(avg(`load5s`)) { `host` = match('cn-hangzhou.172.16.***') } BY `host`", '1m')
  

• 场景 > 仪表板图表可直接下载为 PNG 图片，表格图还可导出为 CSV 文件。
• 日志 > 绑定索引：【字段映射】更改为非必填项。
• 集成/内置视图：模版新增标签管理。
• Service Map 跨工作空间节点样式显示调整。

TrueWatch部署版更新¶

• 管理 > 基本信息 > License 信息：DataKit 数量限制支持按照数据统计范围调整，变更为存活时间 >= 12 小时以上的主机或 DK 数量。
• 支持配置黑名单，自定义选择导入TrueWatch集成、视图模板、监控器模板范围。

OpenAPI 更新¶

• Pipelines 新增/修改：新增 profiling 类型；
• 用户视图新增/修改：支持绑定仪表板配置。

2023 年 4 月 10 日¶

TrueWatch更新¶

• 监控 > 监控器 > 新建：新增【数据断档】、【信息生成】配置区域，以便更好地区分异常数据和数据断档情况。
• 管理：新增系统通知页面，可查看当前账号下的工作空间所有配置的异常状态消息。
• 场景：
• 图表查询：新增 Rollup 函数，该函数同样适用于【指标分析】与【查询工具】；
• 仪表板/用户视图：新增 pin 钉住功能。在当前访问工作空间被授权查看若干其他工作空间数据的前提下，支持将查询其他工作空间数据设为默认选项。
• 系统视图：支持克隆创建为仪表板或者用户视图；
• 自定义查看器：优化搜索模式；非编辑模式下，hover 在【数据范围】即可查看所有筛选条件。
• 查看器 > 快捷筛选：
• 新增【维度分析】按钮，点击后可快速切换到查看器分析模式；
• 支持通过点击外部按钮直接将当前字段添加到显示列/从显示列移除。
• 体验版工作空间 > 导航栏：新增【立即升级】按钮。
• 基础设施 > 容器 > 蜂窝图：新增 CPU 使用率（标准化）和 MEM 使用率（标准化）两种指标填充方式。

TrueWatch部署版更新¶

工作空间管理：新增数据上报限制，帮助利益相关方节约资源等使用成本。

2023 年 3 月 27 日¶

{{ no such element: super_collections.SuperDict object['bbrand_name'] }}更新¶

• 监控：
• 告警策略管理：每条通知规则（包含默认通知和自定义通知）配置新增支持升级通知条件。
• 监控器 > 事件内容：新增自定义高级配置，支持添加关联日志和错误堆栈；
• 主机智能监控：将当前突变展示更改为基于周期以预测的方式进行异常告警，趋势图会展示当前指标及置信区间上下界，超出置信区间的异常会标红展示。
• 场景 > 图表：新增拓扑图。
• APM > 链路详情页 > 服务调用关系：调整为服务拓扑展示，并展示服务与服务之间的调用次数。
• 数据保存策略：Session Replay 的数据保存策略与 RUM 的保存策略保持联动一致，即 RUM 数据保存 3 天，Session Replay 的数据也保存 3 天。
• 查看器：
• 事件查看器 > 基础属性：新增检测指标是否显示配置，缓存到本地，全局适配；
• APM > 错误追踪 > 聚类分析 > 详情页：支持创建异常追踪 Issue；
• RUM > Error > 聚类分析 > 详情页：支持创建异常追踪 Issue；
• RUM > View > 详情页：
  • 性能：新增【所有条目】选项，列出当前 View 下所有关联数据；
  • Fetch/XHR：点击数据行，支持打开对应链路详情页或 Resource 详情页。
• 时间控件：获取 “当前时间” 时，精确到毫秒。
• 管理 > 工单管理：
• 新增评星和评价；
• 已反馈的工单需要调整为 7 天内无客户反馈则自动关单；
• 支持工单列表导出；
• 状态为【已完成】【已撤销】的工单可以进行重启操作；
• 账号注销后其下提交的未关闭状态工单支持自动关闭处理。
• 云市场开通流程整体优化。

TrueWatch部署版更新¶

• 数据保存策略：支持工作空间拥有者配置数据保存策略，且支持自定义输入保存时长。应用场景：
• 指标管理 > 指标集；
• 日志 > 索引 > 新建。
• 用户：支持通过邮箱账号邀请成员。

2023 年 3 月 13 日¶

TrueWatch更新¶

• 监控 > 监控器：监控器类型【组合检测】上线。支持将多个监控器的结果通过表达式组合成一个监控器，最终基于组合后的结果进行告警。
• 场景 > 服务：支持跨工作空间 Service Map 查询。

TrueWatch部署版更新¶

• 管理 > 基本信息：新增 “已用 DK 数量” 显示；
• 管理 > 用户：新增分组页面，基于组可配置关联工作空间及角色，用户可通过组获得对应工作空间的访问权限。

2023 年 3 月 6 日¶

TrueWatch更新¶

• 监控
• 监控器 > 检测频率：新增 Crontab 自定义输入，满足仅需在特定的时间段执行检测的需求；
• 突变检测：新增【最近 1 分钟】、【最近 5 分钟】的检测区间；
• 静默管理：选择静默范围时“事件属性”为非必填项，可根据需要自行配置更细颗粒度的匹配规则。
• DataFlux Func：新增 Function 外部函数。允许第三方用户充分利用 Function 的本地缓存和本地文件管理服务接口编写函数，在工作空间内执行数据分析查询。
• APM > 链路：
• Title 区域 UI 显示优化；
• 针对火焰图、瀑布图、Span 列表超过 1 万的 Span 结果，支持通过偏移设置查看未展示 Span；
• 新增 Error Span 筛选入口；支持输入 Span 对应的资源名称或 Span ID 进行搜索匹配。
• 场景
• 图表：新增桑基图；
• 视图变量：新增选中按钮，勾选后默认全选当前所有值，可按需再反选。
• 账号管理：新增注销入口。
• 查看器：
• UI 显示优化；
• 筛选新增正则匹配 / 反向正则匹配模式；
• Wildcard 筛选和搜索支持左 * 匹配。
• 事件 > 详情页：【告警通知】tab 页 UI 显示优化。

TrueWatch部署版更新¶

• 新增登录方式选择，对登录方式进行统一管理；
• 针对本地账号和单点登录账号新增删除操作。

2023 年 1 月 31 日¶

TrueWatch更新¶

• 监控：
• 智能监控：
  • 主机、日志、应用智能检测频率调整为每 10 分钟执行一次，每执行一次检测计算为 10 次调用费用；
  • 为提升算法精度，日志、应用智能检测采用数据转存的方式，开启一个智能监控后，会生成对应的指标集及指标数据。这一调整会产生额外的时间线，具体数量为当前监控配置的过滤条件所过滤的检测维度数量(service、source) * 检测指标数量，由于没有对监控器的过滤条件进行存储，如果发生监控器过滤条件配置修改的情况，会生成新的等量时间线，所以在修改监控器过滤条件配置当日会有时间线重复计费的情况，修改后次日恢复正常。
• 告警策略管理：
  • 新增自定义通知时间配置，按周期、时间区间细化告警通知配置；
  • 重复告警新增【永久】这一事件选项。
• 监控器
  • 告警配置：支持配置多组告警策略；若配置多个，则 df_monitor_name 与 df_monitor_id 会以多个的形式呈现，并由 ; 分隔开；
  • 联动异常追踪 Issue 改造：新增【事件恢复同步关闭 Issue】开关，当异常事件恢复时，则同步恢复异常追踪 Issue；
  • 监控器列表新增克隆按钮。
• 通知对象管理：新增简单 HTTP 通知类型，直接通过 Webhook 地址接收告警通知；
• 场景：
• 图表：单位新增【货币】选项；高级配置 > 同期对比更改为【同环比】；
• 服务管理 > 资源调用：排行榜新增 TOP / Bottom 数量选择。
• 查看器：显示列设置新增【时间列】开关。
• 付费计划与账单：
• 工作空间锁定弹窗页面新增新建工作空间入口，优化操作体验；
• AWS 注册流程优化。

TrueWatch部署版更新¶

• 支持 LDAP 单点登录；
• 工作空间管理 > 数据存储策略新增自定义选项，范围为<= 1800 天（ 5 年）；其中，指标新增可选项 720 天、1080 天等保存时长；在控制台中设置 > 编辑数据存储策略，修改保存后即可同步更新后台数据存储；
• 用户：支持为用户账号一键配置分配工作空间以及角色；
• 新增控制台审计事件查看入口，可快速查看所有工作空间相关操作审计；
• 新增【管理后台 MFA 认证】。

2023 年 1 月 11 日¶

TrueWatch更新¶

• 日志：
• 新增 BPF 网络日志采集及日志详情页，支持 JSON 格式转化；详情页新增可读的展示模式；
• 新增绑定【关联网络日志】；
• 数据访问：新增批量操作。
• 定时报告：新增可选分享方式【公开分享】或【加密分享】。
• 仪表板：
• 视图变量新增【所有变量值】传参选项；
• 时序图：新增排序逻辑（仅限新引擎），支持针对返回结果进行排序。
• 生成指标：支持批量操作；标准及以上权限成员支持克隆。
• 监控器：
• 通知对象管理：适配新的钉钉机器人，创建时「密钥」选项非必填，快速关联钉钉机器人。
• SLO 扣分逻辑优化。
• 用户访问监测（RUM）：公网 Dataway 支持 ip 转换成地理位置信息。