1.3.1 Release Notes

New Feature

[#1658] Refactor WorkerServer: Using Netty as tcp framework between master and worker, remove zk as task queue, add three choices(Random, RoundRobin, and linear weighted based on cpu and memory) when master choose worker, worker don‘t operate database
[#2584] Support k8s deployment
[#404] Resource center supports directory management
[#1469] Support ambari plugin installation
[#205] Support condition detection type task
[#203] Data synchronization task type, add sqoop support
[#790] Data synchronization task type, add datax support
[#233] Procss definition supports copying, exporting, and importing
[#2501] Add batch export and import workflows feature
[#1460] Improve CI/CD
[#1376] Simplify the deployment, simplify configuration file

BUG FIX

[#3085] after zoom or drag the dag view, cant plcae node correctly and may be disappear
[#3111] Edit the node, after clicking Cancel, the node information is saved
[#3084]remove some standard notice from our NOTICE
[#3032]fix alert.json and metainfo.xml bugs
[#3031]readme.md adds instructions for changing environment variables
[#3058]The task running order in the process instance does not follow the topological order in the process definition
[#3038] remove Conflicting configuration sudo and user
[#3013] fix single resource manager will NPE
[#2728] fix MySQL autoDeserialize bug

1.2.1

New Feature

1. [#1497] DAG automatic layout when use api call process definition create
2. [ #747] Worker server runtime log desensitization
3. [#1635] merge configurations in order to reduce configuration files
4. [#1460] Add CI
5. [#1547] node text edit box supports full-screen magnification
6. [#1646] Add github action to analyze code with sonarcloud
7. [#1367] Add java checkstyle

Enhancement

1. [#184] Use the default workgroup in the page ,Instead of the value id.
2. [#1441] add user success when user name contains '.'.
3. [#839] add Spark Task Component can switch Spark Version.
4. [#1511] Merge frontend and backend tar into one binary tar
5. [#1509] Remove master server and worker server listening port (5566、7788)
6. [#1575] Remove kazoo, simplify deployment
7. [#1300] Add right alignment function in sql email content
8. [#1599] add profile nginx in order to deploy frontend
9. Support Mac local development and debugging

Bug Fixes

1. Solve the problem that the pop-up box cannot be closed when the browser returns
2. Fix [#1399] The wrong field order in logger.info
3. Fix [#1379] SQL task，date parameter need to add explicit type casts
4. Fix [#1477] some tasks would be running all the time when db delayed
5. Fix [#1514] the field queue in table t_ds_user not change with tabke t_ds_queue modify queue_name field
6. Fix [#1768] There are multiple pages of data. After deleting all the data on one page, the data is displayed as empty
7. Fix [#1770] After canceling the file authorization, the running workflow should not obtain resource files from the original tenant directory
8. Fix [#1779] The execution of the SUB_PROCESS task failed first, but eventually succeeded
9. Fix [#1789] Click to view the history, enter the task instance page, the results of the query based on the search conditions are displayed incorrectly
10. Fix [#1810] Workflow instance does not show dependencies
11. Fix [#1816] Add multiple dependencies, the workflow definitions of the first few dependencies read the workflow definition of the last project
12. Fix [#1828] After executing the authorized UDF function, the path of the read resource file is incorrect

1.2.0

New Feature

1. Support postgre sql
2. Change all Chinese names to English
3. Add flink and http task support
4. Cross project dependencies
5. Modify mybatis to mybatisplus, support multy databases.
6. Add export and import definition feaure
7. Github actions ci compile check
8. Add method and parameters comments
9. Add java doc for common module

Enhancement

1. Add license and notice files
2. Move batchDelete Process Define/Instance Outside for transactional
3. Remove space before and after login user name
4. Dockerfile optimization
5. Change mysql-connector-java scope to test
6. Owners and administrators can delete schedule
7. DB page rename and background color modification
8. Add postgre performance monitor
9. Resolve style conflict, recipient cannot tab and value verification
10. Checkbox change background color and env to Chinese
11. Change chinese sql to english
12. Change sqlSessionTemplate singleton and reformat code
13. The value of loadaverage should be two decimal places
14. Delete alert group need delete the relation of user and alert group
15. Remove check resources when delete tenant
16. Check processInstance state before delete worker group
17. Add check user and definitions function when delete tenant
18. Delete before check to avoid KeeperException$NoNodeException

Bug Fixes

1. Fix #1245, make scanCommand transactional
2. Fix ZKWorkerClient not close PathChildrenCache
3. Data type convert error ，email send error bug fix
4. Catch exception transaction method does not take effect to modify
5. Fix the spring transaction not worker bug
6. Task log print worker log bug fix
7. Fix api server debug mode bug
8. The task is abnormal and task is running bug fix
9. Fix bug: tasks queue length error
10. Fix unsuitable error message
11. Fix bug: phone can be empty
12. Fix email error password
13. Fix CheckUtils.checkUserParams method
14. The process cannot be terminated while tasks in the status submit success
15. Fix too many connection in upgrade or create
16. Fix the bug when worker execute task using queue. and remove checking
17. Resole verify udf name error and delete udf error
18. Fix bug: task cannot submit when recovery failover
19. Fix bug: the administrator authorizes the project to ordinary users,but ordinary users cannot see the process definition created by the administrator
20. Fix bug: create dolphinscheduler sql failed

1.1.0 (Not Apache Release)

New features:

1. [EasyScheduler-391] run a process under a specified tenement user
2. [EasyScheduler-288] Feature/qiye_weixin
3. [EasyScheduler-189] Security support such as Kerberos
4. [EasyScheduler-398] Administrator, with tenants (install.sh set default tenant), can create resources, projects and data sources (restricted Have an administrator)
5. [EasyScheduler-293] Click on the parameter selected when running the process, there is no place to view, no save
6. [EasyScheduler-401] Timing is easy to time every second. After the timing is completed, the next trigger time can be displayed on the page.
7. [EasyScheduler-493] add datasource kerberos auth and FAQ modify and add resource upload s3

Enhanced:

1. [EasyScheduler-227] upgrade spring-boot to 2.1.x and spring to 5.x
2. [EasyScheduler-434] The number of worker nodes is inconsistent between zk and mysql
3. [EasyScheduler-435] Verification of the mailbox format
4. [EasyScheduler-441] prohibits running nodes from joining completed node detection
5. [EasyScheduler-400] Home page, queue statistics are not harmonious, command statistics have no data
6. [EasyScheduler-395] For fault-tolerant recovery processes, the status cannot be **Running
7. [EasyScheduler-529] optimize poll task from zookeeper
8. [EasyScheduler-242] worker-server node gets task performance problem
9. [EasyScheduler-352] worker grouping, queue consumption problem
10. [EasyScheduler-461] View data source parameters, need to encrypt account password information
11. [EasyScheduler-396] Dockerfile optimization, and associated Dockerfile and github to achieve automatic mirroring
12. [EasyScheduler-389] service monitor cannot find the change of master/worker
13. [EasyScheduler-511] support recovery process from stop/kill nodes.
14. [EasyScheduler-399] HadoopUtils specifies user actions instead of **Deploying users
15. [EasyScheduler-378] Mailbox regular match
16. [EasyScheduler-625] EasyScheduler call shell "task instance not set host"
17. [EasyScheduler-622] Front-end interface deployment k8s, background deployment big data cluster session error

Repair:

1. [EasyScheduler-394] When the master&worker is deployed on the same machine, if the master&worker service is restarted, the previously scheduled tasks cannot be scheduled.
2. [EasyScheduler-469] Fix naming errors,monitor page
3. [EasyScheduler-392] Feature request: fix email regex check
4. [EasyScheduler-405] Timely modify/add page, start time and end time cannot be the same
5. [EasyScheduler-517] Complement - Sub Workflow - Time Parameters
6. [EasyScheduler-532] Python node does not execute the problem
7. [EasyScheduler-543] optimize datasource connection params safety
8. [EasyScheduler-569] Timed tasks can't really stop
9. [EasyScheduler-463] Email verification does not support very suffixed mailboxes
10. [EasyScheduler-650] Creating a hive data source without a principal will cause the connection to fail
11. [EasyScheduler-641] The cellphone is not supported for 199 telecom segment when create a user
12. [EasyScheduler-627] Different sql node task logs in parallel in the same workflow will be mixed

---

新特性：

1. [EasyScheduler-391] run a process under a specified tenement user
2. [EasyScheduler-288] Feature/qiye_weixin
3. [EasyScheduler-189] Kerberos等安全支持
4. [EasyScheduler-398] 管理员，有租户（install.sh设置默认租户），可以创建资源、项目和数据源（限制有一个管理员）
5. [EasyScheduler-293] 点击运行流程时候选择的参数，没有地方可查看，也没有保存
6. [EasyScheduler-401] 定时很容易定时每秒一次，定时完成以后可以在页面显示一下下次触发时间
7. [EasyScheduler-493] add datasource kerberos auth and FAQ modify and add resource upload s3

增强:

1. [EasyScheduler-227] upgrade spring-boot to 2.1.x and spring to 5.x
2. [EasyScheduler-434] worker节点数量 zk和mysql中不一致
3. [EasyScheduler-435] 邮箱格式的验证
4. [EasyScheduler-441] 禁止运行节点加入已完成节点检测
5. [EasyScheduler-400] 首页页面，队列统计不和谐，命令统计无数据
6. [EasyScheduler-395] 对于容错恢复的流程，状态不能为 **正在运行
7. [EasyScheduler-529] optimize poll task from zookeeper
8. [EasyScheduler-242] worker-server节点获取任务性能问题
9. [EasyScheduler-352] worker 分组, 队列消费问题
10. [EasyScheduler-461] 查看数据源参数，需要加密账号密码信息
11. [EasyScheduler-396] Dockerfile优化，并关联Dockerfile和github实现自动打镜像
12. [EasyScheduler-389] service monitor cannot find the change of master/worker
13. [EasyScheduler-511] support recovery process from stop/kill nodes.
14. [EasyScheduler-399] HadoopUtils指定用户操作，而不是 **部署用户
15. [EasyScheduler-378] Mailbox regular match
16. [EasyScheduler-625] EasyScheduler call shell "task instance not set host"
17. [EasyScheduler-622] Front-end interface deployment k8s, background deployment big data cluster session error

修复：

1. [EasyScheduler-394] master&worker部署在同一台机器上时，如果重启master&worker服务，会导致之前调度的任务无法继续调度
2. [EasyScheduler-469] Fix naming errors,monitor page
3. [EasyScheduler-392] Feature request: fix email regex check
4. [EasyScheduler-405] 定时修改/添加页面，开始时间和结束时间不能相同
5. [EasyScheduler-517] 补数 - 子工作流 - 时间参数
6. [EasyScheduler-532] python节点不执行的问题
7. [EasyScheduler-543] optimize datasource connection params safety
8. [EasyScheduler-569] 定时任务无法真正停止
9. [EasyScheduler-463] 邮箱验证不支持非常见后缀邮箱
10. [EasyScheduler-650] Creating a hive data source without a principal will cause the connection to fail
11. [EasyScheduler-641] The cellphone is not supported for 199 telecom segment when create a user
12. [EasyScheduler-627] Different sql node task logs in parallel in the same workflow will be mixed

1.1.0-preview (Not Apache Release)

New features:

1. [EasyScheduler-391] run a process under a specified tenement user
2. [EasyScheduler-288] Feature/qiye_weixin
3. [EasyScheduler-189] Security support such as Kerberos
4. [EasyScheduler-398] Administrator, with tenants (install.sh set default tenant), can create resources, projects and data sources (restricted Have an administrator)
5. [EasyScheduler-293] Click on the parameter selected when running the process, there is no place to view, no save
6. [EasyScheduler-401] Timing is easy to time every second. After the timing is completed, the next trigger time can be displayed on the page.
7. [EasyScheduler-493] add datasource kerberos auth and FAQ modify and add resource upload s3

Enhanced:

1. [EasyScheduler-227] upgrade spring-boot to 2.1.x and spring to 5.x
2. [EasyScheduler-434] The number of worker nodes is inconsistent between zk and mysql
3. [EasyScheduler-435] Verification of the mailbox format
4. [EasyScheduler-441] prohibits running nodes from joining completed node detection
5. [EasyScheduler-400] Home page, queue statistics are not harmonious, command statistics have no data
6. [EasyScheduler-395] For fault-tolerant recovery processes, the status cannot be **Running
7. [EasyScheduler-529] optimize poll task from zookeeper
8. [EasyScheduler-242] worker-server node gets task performance problem
9. [EasyScheduler-352] worker grouping, queue consumption problem
10. [EasyScheduler-461] View data source parameters, need to encrypt account password information
11. [EasyScheduler-396] Dockerfile optimization, and associated Dockerfile and github to achieve automatic mirroring
12. [EasyScheduler-389] service monitor cannot find the change of master/worker
13. [EasyScheduler-511] support recovery process from stop/kill nodes.
14. [EasyScheduler-399] HadoopUtils specifies user actions instead of **Deploying users

repair:

1. [EasyScheduler-394] When the master&worker is deployed on the same machine, if the master&worker service is restarted, the previously scheduled tasks cannot be scheduled.
2. [EasyScheduler-469] Fix naming errors,monitor page
3. [EasyScheduler-392] Feature request: fix email regex check
4. [EasyScheduler-405] Timely modify/add page, start time and end time cannot be the same
5. [EasyScheduler-517] Complement - Sub Workflow - Time Parameters
6. [EasyScheduler-532] Python node does not execute the problem
7. [EasyScheduler-543] optimize datasource connection params safety
8. [EasyScheduler-569] Timed tasks can't really stop
9. [EasyScheduler-463] Email verification does not support very suffixed mailboxes

---

新特性：

1. [EasyScheduler-391] run a process under a specified tenement user
2. [EasyScheduler-288] Feature/qiye_weixin
3. [EasyScheduler-189] Kerberos等安全支持
4. [EasyScheduler-398] 管理员，有租户（install.sh设置默认租户），可以创建资源、项目和数据源（限制有一个管理员）
5. [EasyScheduler-293] 点击运行流程时候选择的参数，没有地方可查看，也没有保存
6. [EasyScheduler-401] 定时很容易定时每秒一次，定时完成以后可以在页面显示一下下次触发时间
7. [EasyScheduler-493] add datasource kerberos auth and FAQ modify and add resource upload s3

增强:

1. [EasyScheduler-227] upgrade spring-boot to 2.1.x and spring to 5.x
2. [EasyScheduler-434] worker节点数量 zk和mysql中不一致
3. [EasyScheduler-435] 邮箱格式的验证
4. [EasyScheduler-441] 禁止运行节点加入已完成节点检测
5. [EasyScheduler-400] 首页页面，队列统计不和谐，命令统计无数据
6. [EasyScheduler-395] 对于容错恢复的流程，状态不能为 **正在运行
7. [EasyScheduler-529] optimize poll task from zookeeper
8. [EasyScheduler-242] worker-server节点获取任务性能问题
9. [EasyScheduler-352] worker 分组, 队列消费问题
10. [EasyScheduler-461] 查看数据源参数，需要加密账号密码信息
11. [EasyScheduler-396] Dockerfile优化，并关联Dockerfile和github实现自动打镜像
12. [EasyScheduler-389] service monitor cannot find the change of master/worker
13. [EasyScheduler-511] support recovery process from stop/kill nodes.
14. [EasyScheduler-399] HadoopUtils指定用户操作，而不是 **部署用户

修复：

1. [EasyScheduler-394] master&worker部署在同一台机器上时，如果重启master&worker服务，会导致之前调度的任务无法继续调度
2. [EasyScheduler-469] Fix naming errors,monitor page
3. [EasyScheduler-392] Feature request: fix email regex check
4. [EasyScheduler-405] 定时修改/添加页面，开始时间和结束时间不能相同
5. [EasyScheduler-517] 补数 - 子工作流 - 时间参数
6. [EasyScheduler-532] python节点不执行的问题
7. [EasyScheduler-543] optimize datasource connection params safety
8. [EasyScheduler-569] 定时任务无法真正停止
9. [EasyScheduler-463] 邮箱验证不支持非常见后缀邮箱

1.0.5 (Not Apache Release)

Enhanced：

1. [EasyScheduler-597] child process cannot extend father's receivers and cc

Repair：

1. [EasyScheduler-516]The task instance of MR cannot stop in some cases
2. [EasyScheduler-594]soft kill task After the process still exists (parent process child process)
3. [EasyScheduler-569] Timed tasks can't really stop

---

增强:

1. [EasyScheduler-597]child process cannot extend father's receivers and cc

修复：

1. [EasyScheduler-516]The task instance of MR cannot stop in some cases
2. [EasyScheduler-594]soft kill task 后 进程依旧存在(父进程 子进程)
3. [EasyScheduler-569] 定时任务无法真正停止

1.0.4 (Not Apache Release)

Repair:

1. [EasyScheduler-198] The process definition list is sorted according to the timing status and update time.
2. [EasyScheduler-419] Fix online creation of files, hdfs file not created, but returned successfully
3. [[EasyScheduler-481]] (#481) Fix the problem that the job cannot be offline at a time.
4. [EasyScheduler-425] Add kill to its child process when killing a task
5. [EasyScheduler-422] Fixed an issue where the update time and size were not updated when updating the resource file
6. [EasyScheduler-431] Fix the problem of deleting the tenant if the hdfs is not started when deleting the tenant

Enhanced:

1. The message header in the sql task adds support for custom variables.
2. If the mail in the sql task fails, the sql task is failed.
3. Modify the replacement rule of the custom variable in the sql task, support the replacement of multiple single quotes and double quotes
4. When creating a resource file, increase the verification that the resource file already exists on hdfs
5. The shell process exits, the yarn state is not final and waits for judgment.

---

修复：

1. [EasyScheduler-198] 流程定义列表根据定时状态和更新时间进行排序
2. [EasyScheduler-419] 修复在线创建文件，hdfs文件未创建，却返回成功
3. [EasyScheduler-481]修复job不存在定时无法下线的问题
4. [EasyScheduler-425] kill任务时增加对其子进程的kill
5. [EasyScheduler-422] 修复更新资源文件时更新时间和大小未更新的问题
6. [EasyScheduler-431] 修复删除租户时,如果未启动hdfs,则删除租户失败的问题

增强:

1. sql任务中的邮件标题增加了对自定义变量的支持
2. sql任务中的发邮件失败,则此sql任务为失败
3. 修改sql任务中自定义变量的替换规则,支持多个单引号和双引号的替换
4. 创建资源文件时，增加对该资源文件是否在hdfs上已存在的验证
5. shell进程退出，yarn状态非终态等待判断

1.0.3 (Not Apache Release)

1、The process instance is deleted in batches and the corresponding zk queue task is deleted.
2、 Process definition deletion and batch deletion
3、Tenant delete
4、Timed to delete
5、 sql task to add title
6、 sql mail does not show the blank line in front
7、Install.sh join the switch variable, control whether to start the self-start script. Start monitoring self-starting scripts when true.
8、 Resource file rename only modify the description will not report the name has been wrong
9、The sub-parent process global variable override, the sub-process inherits the parent process global variable and can be rewritten
10、 Task dependence increases "today"
11、 Stop and pause to cancel the status check of the process definition
12、 Process instance timeout BUG modification
13、nginx timeout connection problem repair
14、 Front-end one-click deployment optimization, and support ubuntu
15、Project increase process definition statistics and operation process instance statistics
16、The administrator can not generate tokenBUG repair
17、Increase the Master's process tolerance for running, fault tolerant, ready to suspend, and ready to stop.
18、The mail adds SSL protocol and TLS and SSL can be configured to start using
19、Modify the task of scheduling execution, you can resume the process across days, so that the system parameters remain unchanged
20、jar package conflict problem solving
21、 API online documentation

---

1、流程实例批量删除并增加相应zk队列任务删除
2、流程定义删除和批量删除
3、租户删除
4、定时删除
5、sql任务添加title
6、sql邮件不显示前面的空行
7、install.sh加入开关变量，控制是否启动自启动脚本。当为true时启动监控自启动脚本。
8、资源文件重命名只修改描述时不会报名称已存在错误
9、子父流程全局变量覆盖，子流程继承父流程全局变量并可以重写
10、任务依赖增加“今日”
11、停止和暂停取消对流程定义上下线状态检查
12、流程实例超时BUG修改
13、nginx超时连接问题修复
14、前端一键部署优化，并支持ubuntu
15、项目增加流程定义统计和运行流程实例统计
16、管理员不能生成tokenBUG修复
17、增加Master启动对正在运行、容错、准备暂停和准备停止的流程容错
18、邮件增加SSL协议并且TLS和SSL可配置启动使用
19、修改调度执行的任务，可以跨天恢复流程，以致系统参数不变
20、jar包冲突问题解决
21、API在线文档

1.0.2 (Not Apache Release)

1,The schedule is opened to the outside through the token mode, and can be accessed through the api.
2, you can specify the machine (group) where the task runs.
3, task process monitoring and Master, Worker, Zookeeper running status monitoring
4, workflow definition - increase the process timeout alarm
5, the task type supports Oracle, CLICKHOUSE, SQLSERVER, IMPALA
6, Sql task node can independently select CC mail users
7,User Management—Users can bind queues. The user queue level is higher than the tenant queue level. If the user queue is empty, look for the tenant queue.
8, tenant code allows the encoding of pure numbers or underscores
9, Python task can specify the version

---

1,调度通过token方式对外开放接口，可以通过api进行操作访问
2,可以指定任务运行的机器(组)
3,任务流程监控及Master、Worker、Zookeeper运行状态监控
4,工作流定义—增加流程超时报警
5,任务类型支持Oracle、CLICKHOUSE、SQLSERVER、IMPALA
6,Sql任务节点可以独立选取抄送邮件用户
7,用户管理—用户可以绑定队列,用户队列级别高于租户队列级别，如果用户队列为空，则寻找租户队列
8,租户编码允许纯数字或者下划线这种的编码
9,Python任务可指定版本

1.0.1 (Not Apache Release)

[maven-release-plugin] copy for tag 1.0.1