Page History
Table of Contents |
---|
Scheduler maintenance REST API is designed to provide the possibility to interfere when running blueriq on single or multi-nodes environments.
This REST API must be used when:
- Servers shutdowns occur unexpectedly on multi-nodes environments. In this situation the abandoned tasks have to be recovered by calling Cancel All Started Tasks Endpoint.
- Application name and/or version is changed in Blueriq Studio. In this situation the changes are propagated to the database by calling the Update Application Name and Version Endpoint.
These two endpoint are described below:
Cancel All Started Tasks
Situation
Cancel started tasks is the mechanism of recovering the abandoned tasks that were not completed due to unexpected runtime nodes shutdowns.
Blueriq provides two ways for enabling this mechanism:
- Setting the property blueriq.processengine.cancel-started-tasks to true on the process module. This is supported also in previous Blueriq versions and it is designed for single node environments. For more information about process module properties please consult Process Module Properties.
- Using the cancel started tasks endpoint described below. It is designed for multi-nodes environments and overrules the blueriq.processengine.cancel-started-tasks property.
Setting the property
Cancel all started tasks
Description
By calling this endpoint, all the started tasks across all nodes will be reopen. Automatic tasks will also be executed.
Info | ||
---|---|---|
| ||
|
Setting the property blueriq.processengine.cancel-started-tasks to to true when when running in a multi-node environment environment might cause problems because unexpected behavior. For example while the cancellation of started tasks triggered on one runtime node is running, a specific task could be still in progress on another runtime node and it might be reopen.
Cancel all started tasks endpoint was created in order to avoid these types of situations.
Info |
---|
The endpoint overrules the blueriq.processengine.cancel-started-tasks property. |
Usage
The following steps should be followed for using this endpoint:
Access Runtime Swagger Interface:
Info |
---|
When unexpected runtime nodes shutdowns occur:
|
Description
Info | ||
---|---|---|
| ||
|
|
|
|
By calling this endpoint, all started manual tasks across all nodes will be reset to open and all started automatic tasks will executed.
Update Application Name And Version
DescriptionSituation
This endpoint is needed in case you already ran a process enabled project and find out that you have to rename the model name and/or version. Renaming the model in studio doesn't change the cases and jobs references to the model in the database; they still use the old, previous model. The result is that the previously created timers cannot be used anymore, priority algorithm won't be called anymore and tasks with expiration will not expire anymore. To get it back working normally the application name and/or version have to be changed to the new name and/or version (as provided in the studio model) in the database. This can be done by using this endpoint.
Description
By calling one of these endpoint, the application ids for cases matching the application name from the url will be updated with new values. Every job registered to these cases will be updated as wellBy calling this endpoint, the application name, application version or both application name and version will be updated for the cases matching the specified criteria.
Info | ||
---|---|---|
| ||
|
If only new application name is specified in request body, then
Validations
- Current application id consists of application name and application version. Format of the application id is: applicationName:applicationVersion.
- Current application id is mandatory, case id is optional.
If case id is specified, only one case will be updated. If case id is not specified all cases matching the application id will be updated
If only new application version is specified in request body, the application id is updated with the new version only.
If both are specified, the entire application id will be updated
- .
- At least new name or new version must be specified in the request body.
Steps:
- Find all cases matching the application id (from the url). If case id is specified, find the case by it's id.
- Update application id for the cases from step 1, with the new data sent in the request body. Name or version or both can be updated, depending on what is specified in the request body.
- Get the jobs for the cases and update them accordingly.
Request examples:
Update application name
The application ids, for the cases with the current application id .
All cases where application name is "export-TestOldProject:0.0-Trunk", is are updated to "export-TestNewProject:0.0-Trunk".Info icon false PUT http:
//<server>:<port>/<runtime_name>/api/v1/scheduler/maintenance/applications/export-TestOldProject:0.0-Trunk
{
"name" : "export-TestNewProject"
}
Update application version.
The application ids, for the cases with the current application id All cases where application name is "export-TestOldProject:0.0-Trunk", version is are updated to "export-TestOldProject:0.1-Trunk".Info icon false PUT http:
//<server>:<port>/<runtime_name>/api/v1/scheduler/maintenance/applications/export-TestOldProject:0.0-Trunk
{
"version" : "0.1-Trunk"
}
Update application id (both name and version)
All cases where application name is The application ids, for the cases with the current application id "export-TestOldProject:0.0-Trunk", application name is are updated to "export-TestNewProject" and version is updated to ":0.1-Trunk".Info icon false PUT http:
//<server>:<port>/<runtime_name>/api/v1/scheduler/maintenance/applications/export-TestOldProject:0.0-Trunk
{
"name" : "export-TestNewProject",
"version" : "0.1-Trunk"
}
Usage
The following steps should be followed for using this endpoint:
- Update application id for a specific case
- Update application name
The application id, for the case with the current application id "export-TestOldProject:0.0-Trunk" and case id 1, is updated to "export-TestNewProject:0.0-Trunk".
- Update application name
- Make sure that there exists a runtime user with a role that has the permission SCHEDULER_MANAGER.
- Check that blueriq.oauth2.client-id and blueriq.oauth2.secret are defined in application.properties file. If they are missing, please define them.
- Access Runtime Swagger Interface:
Info icon false PUT http:
//<server>:<port>/
<runtime_name>/api/v1
- Fill the username and the password and press Generate Token in order to obtain a new token.
- Expand scheduler.
- Expand PUT
/scheduler/maintenance/applications/
- Specify the old application name you would like to update.
- Specify the body of the request with the new application name, or new application version or both.
- Press Try it out!
Update Process Engine Id for Jobs
Description
By calling this endpoint, the jobs for a specific case will be updated with the new process engine id.export-TestOldProject:0.0-Trunk?caseId=1
{
"name" : "export-TestNewProject"
}
- Update application version
The application id, for the case with the current application id "export-TestOldProject:0.0-Trunk" and case id 1, is updated to "export-TestOldProject:0.1-Trunk".
Info icon false PUT http:
//<server>:<port>/<runtime_name>/api/v1/scheduler/maintenance/
applications/export-TestOldProject:0.0-Trunk?caseId=1
{
"version" : "0.1-Trunk"
}
- Update application id
The application id, for the case with the current application id "export-TestOldProject:0.0-Trunk" and case id 1, is updated to "export-TestNewProject:0.1-Trunk".
Info icon false PUT http:
//<server>:<port>/<runtime_name>/api/v1/scheduler/maintenance/applications/export-TestOldProject:0.0-Trunk?caseId=1
{
"name" : "export-TestNewProject",
"version" : "0.1-Trunk"
}
This functionality does the following:
- Finds the task with the specified task id.
- Takes the case from the task.
- Calls the scheduler to find all jobs for that specific case.
- Updates all the jobs with the new process engine id, taken from the case model (found in step 2).
Usage
The following steps should be followed for using this endpoint:
- Make sure that there exists a runtime user with a role that has the permission SCHEDULER_MANAGER.
- Check that blueriq.oauth2.client-id and blueriq.oauth2.secret are defined in application.properties file. If they are missing, please define them.
Access Runtime Swagger Interface:
Info icon false http://<server>:<port>/<runtime>/api/v1/docs/index.html - Fill the username and the password and press Generate Token in order to obtain a new token.
- Expand scheduler.Expand
- Expand the endpoint you want to use. Available options are:
- PUT /scheduler/maintenance/tasks/cancelStarted
- PUT
- /scheduler/maintenance/applications
- /{
- currentApplicationId}
- Specify the task idrequest parameters.
- Press Try it out!