Scheduler RMI Client API¶
The Denodo Scheduler Client API allows jobs and data sources to be programmatically configured and managed. It also allows to see the list of jobs, start/stop jobs, manage permissions, etc.
To do this, the platform has a facade
com.denodo.scheduler.client.SchedulerManager which allows the
Getting the job list with information on its state (
See a job’s execution reports (
Create, obtain, update, and delete the configuration of a job, and a Data Source (
Enable or disable a job:
Execute a job only once, without taking into account the start and end time for its configuration:
Execute a job with its dependencies (see section Jobs, Start With Dependencies):
Execute a job with state. If the execution of a job gives any errors, executing it again with state will only execute those queries that had errors (see section Jobs, Start with state):
Stop a job from running:
Obtain the types of jobs, and data sources allowed by the system: VDP, CSV, etc.:
Pause the job scheduler. When the scheduler is paused, the jobs will wait until it is resumed before being executed again. This order will only run successfully if no job is being executed:
Resume executing the scheduler when it is paused:
All configuration elements (jobs, and data sources)
are based on the object
which represents a collection of parameters (simple or compound) that
constitute the configuration of the element. The list of parameters and
their specification (data type, mandatoriness, etc.) are given by the
The jobs have an identifier, a description, a type (VDP, VDPCache or VDPIndexer), some associated handlers, a time-based scheduler, and an extraction section which specifies where the data is going to be extracted from (both differ as to the types of data sources that can be used). Jobs of type VDP and VDPIndexer have an export section, where a list of exporters (Scheduler Index, Elasticsearch, JDBC, CSV, custom exporters) is defined. The jobs also have a handlers section, where a list of them is specified (mail, custom handlers) that will be executed after the job export stage. There is also a retry section that enable configuring when and why a job should be reexecuted.
An example is shown below for creating a VDP-type job, assuming that the
data source has been previously created and has the identifier stored in
vdpDatasourceIdentifier variable. This example shows how the
Configuration object for a job is created, using
ParameterStructure objects to specify the job parameters.
Configuration jobConfig = new Configuration(); jobConfig.setDraft(false); jobConfig.setType("job"); jobConfig.setSubType("vdp"); ParameterStructure parameters = new ParameterStructure(); parameters.put(Helper.buildSimpleMonoValuedParam("name", "jobName")); parameters.put(Helper.buildSimpleMonoValuedParam("description", "jobDescription")); parameters.put(Helper.buildSimpleMonoValuedParam("dataSource", vdpDatasourceIdentifier)); parameters.put(Helper.buildSimpleMonoValuedParam("parameterizedQuery", "select * from emp")); Parameter trigger = new CompoundParameter(); trigger.setName("trigger"); Collection<ParameterStructure> params = new ArrayList<ParameterStructure>(); ParameterStructure param = new ParameterStructure(); param.put(Helper.buildSimpleMonoValuedParam("type", "cron")); param.put(Helper.buildSimpleMonoValuedParam("cronExpression", "0 10,44 14 ? 3 WED")); params.add(param); trigger.setValues(params); parameters.put(trigger); jobConfig.setParameters(parameters); schedulerManager.addElement(projectId, jobConfig);
The name of the parameters to use for configuring each type of element
can be obtained from the XML files in
<DENODO_HOME>/metadata/scheduler/elements. They are organized in
subfolders by element type (data source, job, exporter …).
For more information, refer to the Scheduler API Javadoc
and the examples in