Scheduler rules can be applied to optionally limit the types of task that a cluster node can execute. Scheduler rules are applied by updating the SchedulerRules database field, which is located in the rpi_ClusterNodes table in the Pulse operational database. They are configured by updating the field value with a JSON document, as per the following example:
{
"JobRules": [
{
"Type": "Task",
"JobNames": [
"Web events importer",
"Web cache data importer"
]
}
],
"Clients": [
"A31519BE-48F0-452C-9A8B-64E76E199BAA"
]
}
The document's properties are as follows:
-
Type: as per rpi_ExecutionSchedule.Type
-
JobNames: this property can be used to further qualify Type. If Type is set to 'Task', the property can be used to specify specific task name(s), as per rpi_Tasks.TaskName. Else the value specified should be set as per rpi_ExecutionSchedule.Name.
-
Clients: as per rpi_Clients.ClientID
If scheduler rules are applied, a node will only be able to execute work in accordance with the specified restrictions. If a node in a multi-node cluster is configured with scheduler rules, it will be given priority to execute the work to which limited.