VeloDB Manager

VeloDB Manager 24.1.x

VeloDB Manager 24.1.4

December 20, 2024

New features and enhancements

Added retry check optimization for portOpen
Agent does not allow detection to pass when crontab is not configured

Bug fixes

Fixed the issue that the 24.1.3 Cluster inspection was not executed but the task was successful
Fixed the issue that some monitoring charts had no data after the Manager was upgraded to 24.1.3
Fixed the issue of Manager file handle leakage
Fixed the issue of duplicate authorization data in the white version of WebUI

VeloDB Manager 24.1.3

December 12, 2024

New features and enhancements

In cluster settings, support setting FE access proxy address
In advanced configuration of monitoring and alarm service, add configuration of monitoring data retention time and monitoring data path
When installing Mananger/Agent, check whether the dependent tools or scripts are available
Optimize Agent one-click deployment/automatic restart
Manager monitoring adds indicators related to the number of brpc threads
Optimize information about failure to connect to FE cluster
One-click initialization of machine environment
Optimize BE scaling down, Manager execution, support cancellation
Support unbinding of registered unused hosts
Parameter configuration does not restrict parameters such as meta_dir
Manager Agent supports one-click deployment
Support single-node management and control operation retry
Pre-cluster upgrade check whether the metaDir/storageDir recorded by Manager is consistent with the node configuration
Optimize upgrade disk capacity check: scenario where fe_meta and FE are not on the same disk

Bug fixes

Fix the issue that the monitoring chart displays the total capacity of cluster data in the wrong numerical unit
Fix monitoring filter instance not taking effect & The problem of failure to obtain the default number of panels
Fix the issue that the Manager backend database crashes and causes the jump to the initialization page
Fix the issue that the interface returns an error when WebUI adds permissions to the view
Fix the issue of failure to view FE logs
Fix the issue that non-admin users jump to WebUI and report errors
Fix the issue that the permission button cannot be displayed when the WebUI root account connects to 2.0.14 doris
Fix the issue that the peak value of audit log details memory usage is incorrect
Fix the issue that the Manager and Agent heartbeats fail after the cluster creation fails and the retry succeeds
Fix the issue that the inspection does not respond when the cluster status is abnormal
Fix the issue that the Agent is not displayed after upgrading to 24.1.2
Fix the issue that the Manager upgrades the old Doris cluster including the WebUI and reports that the agent does not exist under certain conditions
Fix the issue that the Manager proxy WebUI request timeout is too short
Fix the issue that the Manager writes invalid content to the be.pid file
Fix the issue that the Manager statistical indicators are inconsistent
Fixed the 24.1.3 version to create a 2.0.x version of doris The problem of error message
Fix the issue of restart check metadata_failure_recovery prompt error
Fix the issue of modifying jvm parameters of doris configuration file during upgrade
Fix the issue of monitoring agent default read and write timeout being too small, resulting in query 405 error code
Fix scanning vulnerability

VeloDB Manager 24.1.2

November 4, 2024

Bug Fixes

Fix the issue where tasks stop after retrying a failed full-cluster upgrade.
Support configuration history display and change comparison.
Fix the issue of no data on the monitoring page.
Fix the issue where the Agent's crontab detection did not consider symbolic links.
Optimize the issue where manual restart is required for inconsistent Agent versions.
Add alert metrics and monitoring charts for Stream Load Jobs.
Add monitoring and alerting for missing table replica anomalies.
Optimize configurable parameters that allow modifications.
Optimize the logic for directory cleanup during scaling down.
Optimize the logic for checking the alive status of scaled-down nodes and replica counts.
Verify the presence of installation packages in newly added directories during scaling up.
Support retry for cluster restart tasks.
Check for residual crontab tasks on new nodes added to the takeover cluster.
Optimize the BE decommissioning process.
Optimize the Manager's cleanup of residual directories and files after upgrades.
Automatically and regularly clean up Manager task logs.
Optimize the queuing strategy for asynchronous thread pools in WebUI metadata collection.
Support aliases for WebUI query results.
Fix the issue where null values in WebUI query results are rendered as 0.
Upgrade the tomcat-embed dependency to 9.0.90 to address vulnerabilities in WebUI.

VeloDB Manager 24.1.1

October 21, 2024

Bug Fixes

Fix issues with failure in managing VeloDB 2.0.x clusters.
Resolve issues with abnormal access to WebUI services after Manager upgrades.
Fix missing IP information in alert content and lack of alert conditions during rule replication.
Add health checks to monitoring and resolve nested issues on the monitoring page.
Fix the issue where the notification method for updating regular inspections did not take effect.
Improve error reporting for node checks and refine prompt information for Agent liveliness checks.
Automatically clean up old upgrade directories during Agent upgrades.
Support modifying the log directory through parameter configuration changes.
Add a check for whether the deployment directory is occupied during scaling.
Fix abnormal display of Agent Warning information during node scaling.
Resolve issues with failed sending of regular inspection reports and discrepancies between expected and actual values in inspection data results, even when the results are normal.
Enhance the enterprise edition's WebUI to include the parent level when double-clicking to add table names.
Add pre-upgrade checks, configuration changes, and inspections to verify Agent service availability.
WebUI supports query restrictions for the special "select*" query syntax.

VeloDB Manager 24.1.0

October 8, 2024

New Features

Optimization of the alert module: Policies now support specifying alert levels, importing/exporting, sending notifications upon alert recovery, and enhanced alert information details.
Enhanced upgrade process: Supports incremental (breakpoint) upgrades and rollback in case of upgrade anomalies.
Optimized configuration management: Validates configuration effectiveness and supports interruption of configuration changes.
Integrated webui into manager: Packaged together with manager, eliminating the need for separate deployment.
Webui now supports workload group management, allowing for creation, editing, and viewing of workload groups.
Webui supports import task management, and a new function display has been added to the data section.

VeloDB Manager 24.0.x

VeloDB Manager 24.0.5

September 27, 2024

Bug Fixes

Optimize the Manager's detection logic during database upgrades or FE restarts, and add a time synchronization service check for FE.
Fix the issue with directory permission checks during pre-upgrade.
Fix the problem where the configuration file cannot be found if the deployment directory during scaling does not end with FE.
Fix the issue where the managed cluster cannot find the JDK during an upgrade.
Fix the issue where the BE process does not automatically restart after sending a kill -15 signal.
Fix the problem where the Agent falsely detects BE process death, leading to the execution of the start script.
Fix the issue where some node statuses become inconsistent after management operations in the Manager.
Fix the incorrect unit settings in the File cache configuration.
Fix the issue where, after a failed downsizing attempt, modifying the configuration shows an unknown status while the overview page shows "Running."
Fix the problem where downsizing fails because the Manager shows a BE node as active when it is not actually alive.
Optimize the issue where downsizing of a Mix node is allowed even when only one Mix node exists.
Optimize the failure in deployment caused by the File cache path or storage path being occupied by BE's other paths or duplicate configurations.
Fix the issue where modifying parameters in a managed cluster is unsuccessful.
Fix the issue where alert proxies are not effective.
Fix the incomplete display of error messages during Agent registration.
Fix the issue where the Agent fails to automatically apply strategy configurations.
Optimize the message for inaccurate capacity detection during upgrades.
Fix the issue where the status of a newly created cluster is incorrect (cluster creation fails, but task status shows "In Progress").
Fix the issue where dirty data in the data table causes task anomalies.
Upgrade Spring Web version to 5.3.33.
Fix the problem with external table cache data optimization during cluster creation on the frontend.
Implement various frontend optimizations (such as adding port validation and restrictions, cluster alignment, etc.).

VeloDB Manager 24.0.3

July 17, 2024

Bug Fixes

Support Doris Core 3.0.x (compute-storage coupled mode) deployment.
Add version relationship restrictions between Manager and core versions.
Fix the issue with creating Query Statistics alarm strategies.
Fix the issue of incorrect BE storage path recognition during upgrade directory checks.
Fix the issue of agent auto-upgrade failures.
Optimize invalid status prompts for clusters and nodes.
Fix the issue with incomplete display of upgrade check error messages.

VeloDB Manager 24.0.2

June 14, 2024

Bug Fixes

Fix the issue of incorrect process information retrieval when the manager is hosting.
Resolve the issue of incomplete log viewing due to inconsistent Doris log directories.
Rectify the issue where the status of the restarting node remains as "Restarting".
Fix the incorrect FE/BE Not Alive alert expression.
Address the incorrect value displayed for IO Util in the host interface.
Resolve the issue of the manager initialization page repeatedly re-entering.
Correct the text related to checking if transparent hugepages are disabled.
Optimize the cluster upgrade logic by adjusting the pre-check to before initiating the upgrade operation.
Support filtering label fields for BE nodes.
Improve the error prompts for machine parameter checks, add agent restart prompts, and remove ssh information prompts.

VeloDB Manager 24.0.1

May 6, 2024

Bug Fixes

Fix the issue where BE shrinkage gets stuck.
Fix the problem of garbled content in internal messages.
Optimize error prompts for machine parameter checks, add agent restart prompts, and remove SSH information prompts.
Add a percentage configuration for connections in the alarm strategy.
Fix the issue where a single set of manager services supports deploying multiple agents on a single machine.
Fix the problem where after stopping the FE node, the logs show a normal shutdown but the page still shows it as being in a stopped state.
Fix the inconsistency issue between the BE node PID and the BE.pid record after upgrading the cluster.

VeloDB Manager 24.0.0

April 1, 2024

New Features

Upgrade control to Agent mode, enabling direct HTTP communication between Agent and Server for enhanced security.
Add support for task auditing, providing detailed information such as operation time, operator, and content for each task.
Deployment and scaling support additional computing nodes dedicated to data computation only, separate from data storage.
Support regular inspections, with the ability to configure inspection cycles as needed and send notifications accordingly.
Enhance monitoring module with machine metrics monitoring at the host level.
Include pre-configured alarm templates based on best practices for quick and flexible setup of alarm rules.

VeloDB Manager 23.11.x

VeloDB Manager 23.11.13

Improvements

23.11.x is compatible with 3.0 Doris capacity expansion and contraction ADD node status check

Bug Fixes

Fix the monitoring filter instance not taking effect & the default number of panels failed to obtain the problem
Fix the restart check metadata_failure_recovery prompt error
Fix the restart monitoring component after the Manager upgrade task failed and no log display

VeloDB Manager 23.11.12

November 15, 2024

Improvements

Optimize the manager to clean up the residual directories and files after the upgrade.

Bug Fixes

Fix the issue of no data on the Manager monitoring page
Fix the issue of crontab tasks lacking control components in 23.11.11 deployment Manager
Fix the grafana page nesting problem
Fix the issue of inconsistency between the webui crontab path and the actual deployment path in 23.11.11
Fix the issue of incorrect agentPackage and md5 information caused by the config-tool-configs cache problem
Fix the issue of adding crontab to webui when taking over the cluster
Fix the interface concurrency problem of viewing the deployment manager task in 23.11.12
Fix the issue of not displaying the node list of BE

VeloDB Manager 23.11.11

October 21, 2024

Bug Fixes

Fix the issue where setting the FE load balancing proxy address is ineffective, and causes abnormal issues such as inspections and rolling restarts after setting the proxy address.
Fix the issue where the browser memory gets filled up due to excessive printing of backup data logs generated during BE upgrades.
Fix the issue where inspection data results are normal, but the expected and actual values do not match.
Fix the issue where node status display is inaccurate due to firewall settings.

VeloDB Manager 23.11.10

September 27, 2024

Bug Fixes

Fix the issue where downsizing a single FE removes the crontab for all FEs.
Fix the problem where some prompt messages are incomplete or displayed at the wrong level.
Fix the issue where crontab is not automatically added after a node startup failure.
Fix the criteria for determining invalid node status (switching from API access to querying via show frontends/backends).
Fix the issue where the alert proxy is not effective.

VeloDB Manager 23.11.9

July 17, 2024

Bug Fixes

Support Doris Core 3.0.x (compute-storage coupled mode) deployment.
Add version relationship restrictions between Manager and core versions.
Fix the issue with creating Query Statistics alarm strategies.
Support stopping nodes when the Doris Manager cluster is abnormal.
Optimize invalid status prompts for clusters and nodes.
Fix the issue with incomplete display of upgrade check error messages.

VeloDB Manager 23.11.8

June 7, 2024

Bug Fixes

Fixe the problem of getting wrong process information when manager is hosting
Fixe the problem that the status of the restarted node is always restarting
Fixe the problem that the IO Util in the host interface displays wrong values
Fixe the problem of repeatedly entering the manager initialization page
Fixe the text problem of checking whether the transparent large page is closed
Optimize the cluster upgrade logic and adjusted the pre-check to before the upgrade is initiated
BE node label field supports filtering
Remove avx check when deploying FE

VeloDB Manager 23.11.7

May 6, 2024

Bug Fixes

Add label-based filtering to support Jobs alert policies
Fix the issue of garbled text in site messages on the manager page
Fix the problem where the automatic startup script for the Doris cluster was not working
Add a new configuration for fe connections percentage in alert policies
Fix the issue of slow response when querying the doris_be_cpu chart in Grafana

VeloDB Manager 23.11.6

April 1, 2024

Bug Fixes

Fix the problem of failed alert email notifications
Fix the vulnerability scanning problem with PostgreSQL
Fix the issue of incorrect display of some metric units and alert conditions
Fix the issue of not displaying error information for managed clusters
Resolve the error occurring during the execution of the manager startup script
Add the configuration for FE load balancing proxy addresses

VeloDB Manager 23.11.5

March 18, 2024

Bug Fixes

Upgrade Gragana to address security vulnerabilities identified in older versions during vulnerability scans
Optimize detection logic for obtaining package types during expansion
Update cluster node upgrade checks to include machine time synchronization
Enhance display information for inspections
Optimize logic for data directory utilization during expansion

VeloDB Manager 23.11.4

February 23, 2024

Bug Fixes

Fix the problem of view check judgment when shrinking or upgrading
Fix the problem that the FE node Master role is not displayed during expansion and contraction
Fix the problem that the number of BE nodes is greater than 3 before the node is reduced, and the number of BE nodes is less than 3 after the node is reduced. The problem cannot be reduced.
Optimize Grafana weak password problem
Replace the jdk of the Manager installation package with open jdk
WebUI and Manager optimize standard output file size

VeloDB Manager 23.11.1

December 22, 2023

Bug Fixes

Fix the problem of node role Master information display
Fix the problem of abnormal notification and alarm sending in the site
Fix the configuration problem when the database uses MySQL8
Fix issue with version number selection for managed arm and noavx clusters
Fix the alarm module mailbox information separator bug
Create expansion to remove the strong dependence on the audit loader folder
Optimize deployment, expansion and contraction, and upgrade check logic
Optimize the repair suggestions for some inspection items

VeloDB Manager 23.11.0

December 8, 2023

New Features

Initialization allows specifying case sensitivity when creating a cluster.
Upgrade supports choosing whether to perform backups.
The velodb-doris cluster supports a separate deployment of WebUI.
Service modules support status viewing, restart, and stop functionalities.
Node additions include common Frontend (FE) and Backend (BE) port information, version details, and startup time.

VeloDB Manager 23.10.x

VeloDB Manager 23.10.5

December 1, 2023

Bug Fixes

Optimize the logic for retrying node startup to prevent timeout retries from affecting process startup.
Optimize the logic for scaling down Back-End (BE) nodes, supporting manual DROP of nodes when tablet count is non-zero.
When entering node SSH information, if there is back-and-forth switching and SSH password is ultimately chosen, the frontend may transmit obsolete SSH key information into the SSH password information. Backend storage of this information can lead to control operation failures.
Optimize the logic for decompressing during the version compatibility check phase in cluster upgrade.
Fix Bug where modifying Doris paths does not take effect in the control (ctrl) system.
When accessing the monitoring page, support entering domain information for Grafana in the service configuration.

VeloDB Manager 23.10.1

November 13, 2023

Bug Fixes

Fix Bug where the control service does not release the connection after remote SSH command execution.
Fix Bug where the service configuration update does not reload existing cluster information after monitoring deployment.
Fix Bug in the log service where reading remote files does not clean up unused resources on schedule.

VeloDB Manager 23.10.0

November 10, 2023

New Features

Log Viewing: Supports viewing and querying logs from Front-End (FE) and Back-End (BE) nodes, facilitating offline troubleshooting of cluster issues.
Node List: Added display of WebUI nodes and support for starting and stopping nodes.
Cluster Restart: Cluster restart supports both full restart and rolling restart modes.
Resource Groups: Supports labeling BE nodes, automatically grouping nodes with the same label into a resource group.

VeloDB Manager 23.9.x

VeloDB Manager 23.9.3

November 3, 2023

Bug Fixes

Error Occurs when Modifying Configuration with "file_cache_path" Inclusion
Fix the Issue of Mutual Logouts between WebUI and Manager when logged in on the same machine.

VeloDB Manager 23.9.2

October 27, 2023

Bug Fixes

Check Deployment Directory Space Bug
Modify Configuration File; Backend throws an error if a new key that does not exist in the old configuration file is added.
Fix Bug where some components of the web server are not upgraded during the upgrade process.

VeloDB Manager 23.9.0

October 13, 2023

New Features

Manager：

Service Configuration: Automatically deploy dependent service components through web services. Additionally, it supports modification of service configuration content and self-upgrade of the manager.
User System: Allows the creation of new users and roles, with the ability to configure corresponding roles for users.
Deployment/Scaling with Node Personalized Configuration: New deployments and cluster scaling support customization of parameters such as ports and deployment directories, along with automatic control mode configuration.
Parameter Configuration: Supports custom editing of node configuration files, and individual nodes allow viewing all running parameters.
Node List: Supports the startup and shutdown of Front-End (FE) nodes, with the display of Master node roles.

WebUI：

Data Exploration: Data exploration functionality for log query.
Auto Save: Supports automatic saving of session windows. Saved session windows allow the creation of new folders for organizing sessions.

VeloDB Manager 23.7.x

VeloDB Manager 23.7.8

October 07, 2023

Manager supports upgrading Doris to version 2.0.2 and higher.

VeloDB Manager 23.7.6

September 22, 2023

Bug Fixes

Manager is incompatible with the encoding of CPU/Memory/Network information, preventing successful cluster creation due to database storage issues.
WebUI queries for datetime types now display precision up to milliseconds.

VeloDB Manager 23.7.3

September 01, 2023

Bug Fixes

Support for restarting inactive nodes.
Optimize stop operations to enhance idempotence.
Resolve issues related to systemd leftovers when creating and then taking over through the manager.
Cancel the management of brokers when taking over the cluster.
Compatibility with script execution on Ubuntu systems.

VeloDB Manager 23.7.0

August 10, 2023

New Features

Deploy Cluster : Deploy Apache Doris clusters on physical machines or virtual machines through Doris Manager.
Take Over Cluster : Take over existing Apache Doris clusters into Doris Manager for operation and monitoring.
Cluster Details : View the operational status, details, and connection information of the cluster.
Cluster Scaling : Scale up or down FE and BE nodes within the cluster.
Cluster Upgrade : Conveniently upgrade the cluster version, providing options for full-service stop upgrades and online rolling upgrades. Choose the upgrade method that suits your business scenario.
Cluster Restart : Restart the entire cluster, FE, BE, or specific nodes.
Node Details : View real-time status and machine information of nodes.
Monitoring Alerts : Check predefined monitoring metrics, set up alerts for monitoring metrics, and receive alerts via email, chat software, Webhooks, etc.
Parameter Configuration : Display and configure critical parameters, allowing modifications on a single or batch basis.
Cluster Inspection : Perform a one-click check on machine conditions and cluster operational status, promptly discover and locate performance bottlenecks, and provide repair suggestions.
WebUI : Access the entrance to the cluster's WebUI.

Enterprise Core