ska_pst.lmc.component

This package is for common component classes for PST.LMC.

class ska_pst.lmc.component.MonitorDataHandler(data_store: MonitorDataStore[SubbandMonitorDataType, MonitorDataType], monitor_data_callback: Callable[[MonitorDataType], None])

Generic monitor data handler for subband.

This handler needs to be constructed with a MonitorDataStore that takes subband monitoring data (type S) and can aggregate that data to return a combined monitoring data object (type T).

Since monitoring happens in the background and asynchronously the updates to the data store is guarded by a threading lock.

There should be no need to override any of these methods.

handle_subband_data(subband_id: int, subband_data: SubbandMonitorDataType) → None

Handle subband monitoring data.

This will call the data store’s MonitorDataStore.update_subband() within a write lock. After updating the data store it will call the monitoring callback with the latest monitoring data from the data store.

property monitor_data: MonitorDataType

Get current monitor data.

reset_monitor_data() → None

Reset the monitor data store.

update_monitor_data(notify: bool) → None

Update monitoring data.

This method is used internally as well as by DSP to recalculate monitoring data. To force callback notify must be set to True.

Parameters

notify (bool) – whether to notify callback that there has been an update.

class ska_pst.lmc.component.MonitorDataStore

Generic monitoring data store that handles subband data.

This is an abstract class that other classes should extend to allow following for a contract about how to handle subband data.

property monitor_data: MonitorDataType

Return the current calculated monitoring data.

Implementations of this should aggregate the monitoring data to be what is the current snapshot of data.

Returns

current monitoring data.

reset() → None

Reset the monitoring data state.

update_subband(subband_id: int, subband_data: SubbandMonitorDataType) → None

Update the stored subband data for a given subband id.

This just updates an internal dictionary. Subclasses should only override this method if there is a specific reason to handle a subband, as the monitor_data method should be the main place to handle aggregation.

Parameters

• subband_id – the subband that is being updated.

• subband_data – the data for the current subband.

class ska_pst.lmc.component.PstBaseDevice(*args: Any, **kwargs: Any)

Base class for all the TANGO devices in PST.LMC.

This extends from CspSubElementObsDevice but is also generic in the type of the component manager.

The CspSubElementObsDevice class is used as a base rather than a SKASubarray due to the fact this PST is a software subelement within CSP.

class ConfigureScanCommand(*args: Any, **kwargs: Any)

A class for the ObsDevice ConfigureScan command.

This overrides the constructor to ensure that the correct validator is used. The base CSP ObsDevice ConfigureScanCommand doesn’t allow passing through a validator.

class ScanCommand(*args: Any, **kwargs: Any)

A class for the PST’s ConfigureScan command.

always_executed_hook() → None

Execute call before any TANGO command is executed.

property beam_id: int

Get the ID of the beam this device belongs to.

property component_manager: T

Get component manager.

Overrides the super class property of component_manager to be typesafe.

Returns

the component manager casted to type T.

create_component_manager() → T

Create and return a component manager for this device.

Raises

NotImplementedError – for no implementation

delete_device() → None

Delete resources allocated in init_device.

This method allows for any memory or other resources allocated in the init_device method to be released. This method is called by the device destructor and by the device Init command.

property device_name: str

Get the name of the device.

This is the relative device name (e.g. low_psi/beam/01) and not the FQDN which can include the Tango DB in a URL.

property facility: TelescopeFacilityEnum

Get the facility that this device is being used for.

handle_attribute_value_update(attribute_name: str, value: Any) → None

Handle update of a device attribute value.

Parameters

• attribute_name (str) – the name of the attribute to update.

• value (Any) – the new value of the attribute to update to.

handle_communication_state_change(communication_state: ska_control_model.CommunicationStatus) → None

Handle a change in device’s communication state.

This just calls the SKABaseDevice._communication_state_changed method

handle_component_state_change(*args: Any, **kwargs: Any) → None

Handle change in this device’s state.

This overrides the PstDeviceInterface and calls _component_state_changed on this class.

handle_fault(fault_msg: str) → None

Handle putting the device into a fault state.

init_command_objects() → None

Set up the command objects.

init_device() → None

Initialise the attributes and properties of the device.

This overrides the SKABaseDevice.

property obs_state: ska_control_model.ObsState

Get the current observing state of the device.

Returns

the current observing state of the device.

Return type

ObsState

set_logging_level(value: ska_control_model.LoggingLevel) → None

Set the logging level for the device.

Both the Python logger and the TANGO logger are updated.

This calls SKABaseDevice.set_logging_level() then delegates to the component manager to perform necessary updates. This is used by having the BEAM.MGMT to update the subordinate Tango devices, which in turn update the core apps.

Parameters

value – Logging level for logger

Raises

LoggingLevelError – for invalid value

simulationMode(value: ska_control_model.SimulationMode) → None

Set the simulation mode.

Parameters

value – The simulation mode, as a SimulationMode value

property subsystem_id: str

Get the sub-system id where device is deployed.

update_health_state(health_state: ska_control_model.HealthState) → None

Update the health state of the device.

This delegates to the base class _update_health_state

Parameters

health_state (HealthState) – the health state of the beam to update to.

class ska_pst.lmc.component.PstComponentManager(*args: Any, **kwargs: Any)

Base Component Manager for the PST.LMC. subsystem.

This base class is used to provide the common functionality of the PST TANGO components, such as providing the the communication with processes that are running (i.e. RECV, DSP, or SMRB).

This class also helps abstract away calling out to whether we’re using a simulated process or a real subprocess.

This component manager extends from the CspObsComponentManager. For more details about this check the CSP obs component manager docs.

property beam_id: int

Return the beam id for the current component.

This value is set during the construction of the component manager, and is injected from the DeviceID property of the TANGO device.

property config_id: str

Return the configuration id.

configure_beam(task_callback: Optional[Callable[[...], Any]] = None, **kwargs: Any) → Tuple[ska_tango_base.executor.TaskStatus, str]

Configure the beam specific configuration of the component.

Parameters

• task_callback (Callback) – callback for background processing to update device status.

• kwargs (dict) – configuration for beam

configure_scan(task_callback: Optional[Callable[[...], Any]] = None, **kwargs: Any) → Tuple[ska_tango_base.executor.TaskStatus, str]

Configure the component for a scan.

Parameters

• task_callback (Callback) – callback for background processing to update device status.

• kwargs (dict) – the configuration to be configured

deconfigure_beam(task_callback: Optional[Callable[[...], Any]] = None) → Tuple[ska_tango_base.executor.TaskStatus, str]

Deconfigure the component’s beam configuration.

This will release all the resources associated with the component, including the SMRBs.

Parameters

task_callback (Callback) – callback for background processing to update device status.

deconfigure_scan(task_callback: Optional[Callable[[...], Any]] = None) → Tuple[ska_tango_base.executor.TaskStatus, str]

Deconfigure this component for current scan configuration.

Parameters

task_callback (Callback) – callback for background processing to update device status.

property device_name: str

Get the name of the current device.

go_to_fault(fault_msg: str, task_callback: Optional[Callable[[...], Any]] = None) → Tuple[ska_tango_base.executor.TaskStatus, str]

Set the component into a FAULT state.

For BEAM this will make the sub-devices be put into a FAULT state. For API backed component managers it is expected that the service backing that API should be put into a FAULT state.

obsreset(task_callback: Optional[Callable[[...], Any]] = None) → Tuple[ska_tango_base.executor.TaskStatus, str]

Reset the component to unconfigured but do not release resources.

Parameters

task_callback (Callback) – callback for background processing to update device status.

property scan_id: int

Return the scan id.

set_logging_level(log_level: ska_control_model.LoggingLevel) → None

Set LoggingLevel.

Parameters

log_level – The required TANGO LoggingLevel

Returns

None.

property simulation_mode: ska_control_model.SimulationMode

Get value of simulation mode state.

Returns

current simulation mode state.

start_communicating() → None

Establish communication with the component, then start monitoring.

This is the place to do things like:

• Initiate a connection to the component (if your communication is connection-oriented)

• Subscribe to component events (if using “pull” model)

• Start a polling loop to monitor the component (if using a “push” model)

stop_communicating() → None

Cease monitoring the component, and break off all communication with it.

For example,

• If you are communicating over a connection, disconnect.

• If you have subscribed to events, unsubscribe.

• If you are running a polling loop, stop it.

class ska_pst.lmc.component.PstDeviceInterface

A purely abstract class to be implemented by TANGO device classes.

This class is used to abstract away any TANGO functionality that component managers need to callback to the TANGO device and in turn allow passing of the TANGO device itself to the component manager but is abstracted. This class itself can be extended for a particular device/component_manager combination where there is a need for more specific functionality but without the need of exposing a callback.

By being an abstract class this can be mocked to be used in testing.

property beam_id: int

Get the beam id for the current device.

property device_name: str

Get the name of the device.

property facility: TelescopeFacilityEnum

Get the facility that this device is being used for.

handle_attribute_value_update(attribute_name: str, value: Any) → None

Handle update of a device attribute value.

Parameters

• attribute_name (str) – the name of the attribute to update.

• value (Any) – the new value of the attribute to update to.

handle_communication_state_change(communication_state: ska_control_model.CommunicationStatus) → None

Handle a change in device’s communication state.

handle_component_state_change(*args: Any, **kwargs: Any) → None

Handle a change in the component’s state.

handle_fault(fault_msg: str) → None

Handle device going into a fault state.

property health_check_interval: int

Get the health check interval, in milliseconds.

property monitoring_polling_rate_ms: int

Get the monitoring polling rate, in milliseconds.

property obs_state: ska_control_model.ObsState

Get the current observing state of the device.

Returns

the current observing state of the device.

Return type

ObsState

property subsystem_id: str

Get the sub-system id where device is deployed.

update_health_state(health_state: ska_control_model.HealthState) → None

Update the health state of device.

Parameters

health_state (HealthState) – the new health state of the TANGO device.

class ska_pst.lmc.component.PstGrpcLmcClient(client_id: str, endpoint: str, default_timeout: float = 60.0, logger: Optional[Logger] = None, **kwargs: Any)

The client API that connects to a remote gRPC service.

This client is a wrapper around the PstLmcServiceStub that is generated from the gRPC/Protobuf bindings.

Once fully implemented this class will be able to be used by any of the LMC components PstProcessApi implementations.

abort(timeout: Optional[float] = None, **kwargs: Any) → None

Abort scanning.

This method is to be used by the LMC device that needs to abort a long running action, in particular scan. The ObsState model allows for this to be called if in IDLE (resources assigned), CONFIGURING (configuring a scan), READY (configured for a scan but not scanning), SCANNING (a scan is running), or RESETTING (is trying to reset from ABORTED/FAULT state).

After this call the state of the service should be ABORTED.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

configure_beam(request: ska_pst.grpc.lmc.ska_pst_lmc_pb2.ConfigureBeamRequest, timeout: Optional[float] = None, **kwargs: Any) → None

Call configure_beam on remote gRPC service.

Parameters

• request (ConfigureBeamRequest) – the configure beam request object.

• timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

configure_scan(request: ska_pst.grpc.lmc.ska_pst_lmc_pb2.ConfigureScanRequest, timeout: Optional[float] = None) → None

Call configure_scan on remote gRPC service.

Parameters

• request (ConfigureScanRequest) – the configure can request object for the remote gRPC service.

• timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

connect(timeout: Optional[float] = None, **kwargs: Any) → None

Connect client to the remote gRPC service.

This is used to let the server know that a client has connected.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

deconfigure_beam(timeout: Optional[float] = None, **kwargs: Any) → None

Call deconfigure_beam on remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

deconfigure_scan(timeout: Optional[float] = None, **kwargs: Any) → None

Call deconfigure_scan on remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

get_beam_configuration(timeout: Optional[float] = None) → ska_pst.grpc.lmc.ska_pst_lmc_pb2.GetBeamConfigurationResponse

Call get_beam_configuration on remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

Returns

the current beam configuration on the remote gRPC service.

Return type

GetBeamConfigurationResponse

get_env(timeout: Optional[float] = None, **kwargs: Any) → dict

Get the environment values from the remote gRPC service.

This will map the Protobuf EnvVal objects to the appropriate Python types.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

Returns

the environment values from the remote gRPC service.

Return type

dict

get_log_level(timeout: Optional[float] = None, **kwargs: Any) → ska_pst.grpc.lmc.ska_pst_lmc_pb2.LogLevel

Get the LogLevel of the remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

Returns

The current LogLevel of the remote gRPC service.

Return type

LogLevel

get_scan_configuration(timeout: Optional[float] = None) → ska_pst.grpc.lmc.ska_pst_lmc_pb2.GetScanConfigurationResponse

Call get_scan_configuration on remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

Returns

the current scan configuration on the remote gRPC service

Return type

GetScanConfigurationResponse

get_state(timeout: Optional[float] = None, **kwargs: Any) → ska_control_model.ObsState

Call get_state on remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

Returns

the current observation state of the remote gRPC service

Return type

ObsState

go_to_fault(timeout: Optional[float] = None, **kwargs: Any) → None

Put the gRPC service in to a FAULT state.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

monitor(abort_event: Event, *, polling_rate: int = 5000) → Generator[ska_pst.grpc.lmc.ska_pst_lmc_pb2.MonitorResponse, None, None]

Call monitor on remote gRPC service.

This method is not decorated with the grpc_request as it handles timeout differently but this does handle the gRPC errors just like the grpc_request decorator does.

Parameters

• abort_event (Event) – a threading.Event that can be used to signal to stop monitoring.

• polling_rate (int, optional) – the rate, in milliseconds, at which the monitoring should poll. The default value is 5000ms (i.e. 5 seconds).

Yield

the current monitoring state from the remote gRPC service

Return type

Generator[MonitorResponse, None, None]

perform_health_check(abort_event: Event, *, health_check_interval: int = 1000) → Generator[ska_pst.grpc.lmc.ska_pst_lmc_pb2.HealthCheckResponse, None, None]

Perform health check on remote gRPC service.

This method is not decorated with the grpc_request as it handles timeout differently but this does handle the gRPC errors just like the grpc_request decorator does.

Parameters

• abort_event (Event) – a threading.Event that can be used to signal to stop performing the health check of the remote gRPC service.

• health_check_interval (int, optional) – the interval, in milliseconds, to expect a response back from the remote gRPC service, defaults to 1000

Yield

the current health check response

Return type

Generator[HealthCheckResponse, None, None]

reset(timeout: Optional[float] = None, **kwargs: Any) → None

Reset service.

This method is to be used by the LMC device that is currently in an ABORTED or FAULT state to reset the service. After this call the state of the service should be in IDLE (resources assigned and not configured for a scan).

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

restart(timeout: Optional[float] = None) → None

Restart service.

This method is to be used by the LMC device to restart a service regardless of its ObsState. After this call the state of the service should be in IDLE (resources assigned and not configured for a scan).

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

set_log_level(request: ska_pst.grpc.lmc.ska_pst_lmc_pb2.SetLogLevelRequest, timeout: Optional[float] = None, **kwargs: Any) → None

Set the LogLevel of the remote gRPC service.

Parameters

• request (SetLogLevelRequest) – The request containing LogLevel to be set on the remote gRPC service.

• timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

start_scan(request: ska_pst.grpc.lmc.ska_pst_lmc_pb2.StartScanRequest, timeout: Optional[float] = None, **kwargs: Any) → None

Call start_scan on remote gRPC service.

Parameters

• request (StartScanRequest) – the start scan request to send to the remote gRPC service

• timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

stop_scan(timeout: Optional[float] = None, **kwargs: Any) → None

Call stop_scan on remote gRPC service.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to self.default_timeout

class ska_pst.lmc.component.PstObsStateMachine(callback: Optional[Callable] = None, initial_state: str = 'IDLE', **extra_kwargs: Any)

State machine for observation state.

This class to be used in place of the ska_control_model._ObsStateMachine state machine to allow for the specifics of the state machine needed for the PST Process sub-devices (i.e. SMRB.MGMT, DSP.MGMT, RECV.MGMT, etc)

This is based off the SKA ObsStateModel

The full list of supported states are:

• IDLE: PST has not been configured for a scan

• CONFIGURING_IDLE: PST is in the process of configuring a scan;

  it is currently performing a configure scan operation

• CONFIGURING_READY: PST is in the process of configuring a scan; and is currently finishing the configure scan operation

• READY: PST is configured for a scan

• SCANNING: PST is scanning

• ABORTING: PST is aborting

• ABORTED: PST has aborted

• RESETTING: PST is resetting to IDLE

• FAULT: PST has faulted and needs to be reset

A diagram of the state machine is shown below. Reflexive transitions and transitions to FAULT obs state are omitted to simplify the diagram.

Diagram of PST subdevice state machine

class ska_pst.lmc.component.PstObsStateModel(*args: Any, **kwargs: Any)

Implements the observation state model for a PST Device.

The implemented states are:

• IDLE: the device is unconfigured.

• CONFIGURING: transitional state to report device configuration is in progress.

• READY: the device is configured and is ready to perform observations

• SCANNING: the device is performing the observation.

• ABORTING: the device is processing an abort.

• ABORTED: the device has completed the abort request.

• RESETTING: the device is resetting all the device back to IDLE

• FAULT: the device component has experienced an error from which it can be recovered only via manual intervention invoking a reset command that force the device to the base state (IDLE).

class ska_pst.lmc.component.PstProcessApi(*args: Any, **kwargs: Any)

Abstract class for the API of the PST.LMC processes like RECV, SMRB, etc.

abort() → None

Abort a scan.

configure_beam(configuration: dict) → None

Configure beam for service.

Parameters

configuration – Dictionary of resources to allocate.

configure_scan(configuration: dict) → None

Configure a scan.

Parameters

configuration – the scan configuration for the device.

connect() → None

Connect to the external process.

deconfigure_beam() → None

Deconfigure beam to release all resources.

deconfigure_scan() → None

Deconfigure a scan.

disconnect() → None

Disconnect from the external process.

get_env() → dict

Get the environment properties for the service.

go_to_fault() → None

Set remote service in a FAULT state.

This doesn’t take a callback as we want a synchronous call.

monitor(subband_monitor_data_callback: Callable[[...], None], polling_rate: int = 5000, monitor_abort_event: Optional[Event] = None) → None

Monitor data of remote service.

This needs to be implemented as a background task

Parameters

• subband_monitor_data_callback – callback to use when there is an update of the sub-band monitor data.

• polling_rate – the rate, in milliseconds, at which the monitoring should poll. The default value is 5000ms (i.e. 5 seconds).

• monitor_abort_event – a threading.Event that can be used to signal to stop monitoring. If not set then the background task will create one.

perform_health_check(health_check_handler: HealthCheckHandler, health_check_interval: int = 1000, health_check_abort_event: Optional[Event] = None) → None

Perform health check of a process in the background.

Parameters

• health_check_handler (HealthCheckHandler) – an object that implements the HealthCheckHandler protocol. Any health check state object that is returned from the service is delegated to this handler to be handled.

• health_check_interval (int, optional) – the interval, in milliseconds, at which health check should be perform, defaults to 1000 (i.e. 1 second).

• health_check_abort_event (Optional[threading.Event], optional) – a threading primitive to be used to stop the health check by an external mechanism, defaults to None

reset() → None

Reset the component.

restart() → None

Restart the component.

set_log_level(log_level: ska_control_model.LoggingLevel) → None

Set the LogLevel of the service.

Parameters

log_level – The required TANGO LoggingLevel

Returns

None.

start_scan(scan_id: int, **kwargs: Any) → None

Start a scan.

Parameters

• scan_id (int) – the ID for the scan.

• kwargs (dict) – additional arguments, needed to allow for future proofing of scan request coming from TM / CSP.

stop_health_check() → None

Stop performing health check of service.

stop_monitoring() → None

Stop background monitoring.

stop_scan() → None

Stop a scan.

validate_configure_beam(configuration: dict) → None

Validate a configure beam for service.

Parameters

configuration – Dictionary of resources to allocate.

Raises

ValidationError – if there an issue validating the request. The error message contains the details.

validate_configure_scan(configuration: dict) → None

Validate a configure_scan request.

Parameters

configuration – the scan configuration for the device.

Raises

ValidationError – if there an issue validating the request. The error message contains the details.

class ska_pst.lmc.component.PstProcessApiGrpc(*args: Any, **kwargs: Any)

Helper class to be used by subclasses of PstProcessApi that use gRPC.

This class should be added as a parent class of gRPC client APIs. Common logic of methods can be refactored to this class. This also means that requests that have empty request messages can be handled by this class specifically. Where request parameters need to be converted to the appropriate protobuf message, then subclasses of this class need to implement the _get_<method_name>_request.

For monitoring the subclasses must implement the _handle_monitor_response method.

abort() → None

Abort a scan.

configure_beam(configuration: dict, timeout: Optional[float] = None) → None

Configure the beam with the resources defined in configuration.

Parameters

• configuration (dict) – Dictionary of resources to allocate.

• timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to default_timeout attribute of the gRPC API client.

Raises

BaseGrpcException – if there was a un-recoverable server server error.

configure_scan(configuration: dict) → None

Configure a scan.

Parameters

configuration (dict) – the configuration for the scan.

connect() → None

Connect to the remote gRPC service.

deconfigure_beam() → None

Deconfigure the beam, releasing all resources.

deconfigure_scan() → None

Deconfigure a scan.

disconnect() → None

Disconnect from the external process.

This will ensure any monitoring background task has stopped.

get_env() → dict

Get the environment properties from the remote gRPC service.

get_log_level() → ska_control_model.LoggingLevel

Get the LogLevel of the remote gRPC service.

go_to_fault() → None

Put remote service into FAULT state.

This is used to put the remote service into a FAULT state to match the status of the LMC component.

monitor

perform_health_check

reset() → None

Reset service.

restart() → None

Restart service.

set_log_level(log_level: ska_control_model.LoggingLevel) → None

Set the LogLevel of the remote gRPC service.

Parameters

log_level – The required TANGO LoggingLevel.

Returns

None.

start_scan(scan_id: int, **kwargs: Any) → None

Start scanning.

Parameters

• scan_id (int) – the ID for the scan.

• kwargs (dict) – additional arguments, needed to allow for future proofing of scan request coming from TM / CSP.

stop_health_check() → None

Stop performing health check of service.

stop_monitoring() → None

Stop background monitoring.

stop_scan(timeout: float = 60.0) → None

End a scan.

This will call out to the remote service to end a scan. It will also stop monitoring as monitoring is only valid if the service is in a scan.

Parameters

timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to default_timeout attribute of the gRPC API client.

validate_configure_beam(configuration: dict, timeout: Optional[float] = None) → None

Validate configuration for a configure_beam request.

Parameters

• configuration (dict) – Dictionary of resources to allocate.

• timeout (float | None, optional) – the timeout, in seconds, for the request, defaults to default_timeout attribute of the gRPC API client.

Raises

• ValidationError – if there is an issue validating the request. The error message contains the details.

• TimeoutException – if the request took longer than expected.

validate_configure_scan(configuration: dict) → None

Validate a configure_scan request.

Parameters

configuration (dict) – the configuration for the scan.

Raises

ValidationError – if there is an issue validating the request. The error message contains the details.

class ska_pst.lmc.component.PstProcessApiSimulator(*args: Any, **kwargs: Any)

Abstract class for the Simulated API of the PST.LMC processes like RECV, SMRB, etc.

abort() → None

Abort a scan.

configure_beam(configuration: dict) → None

Configure the beam.

Parameters

configuration – dictionary of parameters to be configured and their requested values

configure_scan(configuration: dict) → None

Configure a scan.

Parameters

configuration – the configuration for the scan.

connect() → None

Connect to the external process.

deconfigure_beam() → None

Deconfigure the beam.

The default implementation of this does nothing.

deconfigure_scan() → None

Deconfigure a scan.

disconnect() → None

Disconnect from the external process.

get_env() → dict

Get the environment properties for the service.

go_to_fault() → None

Set simulator into a FAULT state.

If simulator is scanning then stop scanning.

monitor

perform_health_check(health_check_handler: HealthCheckHandler, health_check_interval: int = 1000, health_check_abort_event: Optional[Event] = None) → None

Perform health check of a process in the background.

The simulator process doesn’t perform health check. This may change in the future to simulate errors.

Parameters

• health_check_handler (HealthCheckHandler) – an object that implements the HealthCheckHandler protocol. Any health check state object that is returned from the service is delegated to this handler to be handled.

• health_check_interval (int, optional) – the interval, in milliseconds, at which health check should be perform, defaults to 1000 (i.e. 1 second).

• health_check_abort_event (Optional[threading.Event], optional) – a threading primitive to be used to stop the health check by an external mechanism, defaults to None

reset() → None

Reset service when in ABORTED / FAULT state.

restart() → None

Restart service in any state.

set_log_level(log_level: ska_control_model.LoggingLevel) → None

Set simulator LoggingLevel of the PST.LMC processes like RECV, SMRB, etc.

Parameters

log_level – The required TANGO LoggingLevel

Returns

None.

start_scan(scan_id: int, **kwargs: Any) → None

Start a scan.

Parameters

• scan_id (int) – the ID for the scan.

• kwargs (dict) – additional arguments, needed to allow for future proofing of scan request coming from TM / CSP.

stop_health_check() → None

Stop performing health check of service.

As the simulator API doesn’t perform health check this method does nothing.

stop_monitoring() → None

Stop the monitoring background thread by setting event.

stop_scan() → None

Stop a scan.

validate_configure_beam(configuration: dict) → None

Validate configure beam request.

Parameters

configuration (dict) – Dictionary of resources to allocate.

Raises

ValidationError – if there is an issue validating the request. The error message contains the details.

validate_configure_scan(configuration: dict) → None

Validate configure scan request.

class ska_pst.lmc.component.PstProcessApiSubcomponentManager(*args: Any, **kwargs: Any)

A base subcomponent Manager for the PST.LMC that uses an API.

This extends from the PstBaseSubcomponentManager

connect() → None

Establish connection to API subcomponent.

deconfigure_beam() → None

Deconfigure the subcomponent’s beam configuration.

This will release all the resources associated with the subcomponent, including the SMRBs.

disconnect() → None

Disconnect from API subcomponent.

get_env() → dict

Get the environment properties for the service.

handle_health_check_state(state: HealthCheckState, **kwargs: Any) → None

Handle a health check state from the subcomponent.

This is an implementation of the ska_pst.lmc.health_check.HealthCheckHandler protocol.

Parameters

state (HealthCheckState) – the current health check state of subcomponent.

property health_check_state_mismatch_count: int

Get the number of times that the health check state was not the expected state.

property monitor_data: MonitorDataType

Get the current monitoring data.

property obs_state: ska_control_model.ObsState

Get the current observing state of sub-component.

Returns

the current observing state of sub-component.

Return type

ObsState

start_health_check() → None

Start performing health check on subcomponent in the background.

The background health check processing is delegated to the process API.

stop_health_check() → None

Stop background health check on subcomponent.

validate_configure_scan(configuration: dict) → None

Validate configure scan request with the specific configuration of the subcomponent.

Note this is for the whole ConfigureScan request for a PST BEAM subcomponent, which includes checking both the beam and scan configuration is correct. This is due to the fact that a client of BEAM.MGMT only exposes a ConfigureScan request as it’s an Obs device.

Parameters

configuration (dict) – the configuration to validate.

class ska_pst.lmc.component.PstSimulator(*args: Any, **kwargs: Any)

A base class for simulators used when the BEAM is in simulation mode.

A simulator class is needed for each subcomponent. This class is generic over the subband monitoring data type and a monitoring data store.

This is an abstract class and sub-classes of this are expected to implement the _update() method.

abort() → None

Tell the component to abort whatever it was doing.

configure_scan(configuration: dict) → None

Simulate configuring a scan.

Parameters

configuration (dict) – the configuration to be configured

deconfigure_scan() → None

Simulate deconfiguring of a scan.

get_subband_data() → Dict[int, SubbandMonitorDataType]

Get simulated subband data.

start_scan(**kwargs: Any) → None

Simulate start scanning.

Param

the scan arguments.

stop_scan() → None

Simulate stop scanning.

class ska_pst.lmc.component.PstSubcomponentManager(*args: Any, **kwargs: Any)

An abstract base sub component manager.

This class is used as a base class for all the PST subcomponents that are managed by the PST BEAM.MGMT TANGO device. For specific components such as RECV.CORE, SMRB.CORE, DSP.DISK, DSP.FT, etc, the sub-component managers should extend from the PstProcessApiSubcomponentManager which uses a process API to manage the core application.

In the case of the DSP sub-components an umbrella DSP subcomponent manager should be used to orchestrate the requests going to a specific sub-component depending on the scan configuration and the processing mode.

abort() → None

Abort current process.

The only long lived process for API based devices is that of SCANNING. However, if another system fails this can be used to put all the subsystems into an ABORTED state.

property beam_id: int

Get beam ID for the current subcomponent.

configure_beam(configuration: dict) → None

Configure the beam specific configuration of the component.

Parameters

configuration (dict) – configuration for beam

configure_scan(configuration: dict) → None

Configure the component for a scan.

Parameters

configuration (dict) – the configuration to be configured

connect() → None

Establish connection to API component.

deconfigure_beam() → None

Deconfigure the component’s beam configuration.

This will release all the resources associated with the component, including the SMRBs.

deconfigure_scan() → None

Deconfigure this component for current scan configuration.

Parameters

task_callback (Callback) – callback for background processing to update device status.

disconnect() → None

Establish connection to API component.

end_scan() → None

Stop scanning.

property fault_msg: str | None

Get the current fault message.

go_to_fault(fault_msg: str) → None

Set the component into a FAULT state.

For BEAM this will make the sub-devices be put into a FAULT state. For API backed component managers it is expected that the service backing that API should be put into a FAULT state.

property health_check_interval: int

Get the current health check interval, in milliseconds.

Returns

the current health check interval, in milliseconds.

Return type

int

property health_state: ska_control_model.HealthState

Get the current health state of the sub-component.

Returns

the current health state of the sub-component.

Return type

HealthState

property monitoring_polling_rate_ms: int

Get the current monitoring polling rate, in milliseconds.

property obs_state: ska_control_model.ObsState

Get the current observing state of sub-component.

Returns

the current observing state of sub-component.

Return type

ObsState

obsreset() → None

Reset service.

This is used to reset a service in ABORTED or FAULT states back to an EMPTY state. This will deconfigure a scan and beam.

reset() → None

Restart service.

This is used to restart a service regardless of state.

reset_monitoring() → None

Stop monitoring and reset monitoring data.

restart_health_check() → None

Restart the health check background process for the current subcomponent.

scan(scan_id: int, **kwargs: Any) → None

Start scanning.

The kwargs of this method is scan request. By using the kwargs allow for forward compatibility of accepting other parameters for the starting of the scan.

Parameters

• scan_id (int) – the scan ID

• kwargs (dict) – scan request as a dict

set_logging_level(log_level: ska_control_model.LoggingLevel) → None

Set LoggingLevel.

Parameters

log_level – The required TANGO LoggingLevel

Returns

None.

property simulation_mode: ska_control_model.SimulationMode

Get value of simulation mode state.

Returns

current simulation mode state.

start_health_check() → None

Start performing health check on subcomponent.

stop_health_check() → None

Stop background health check on subcomponent.

validate_configure_scan(configuration: dict) → None

Validate configure scan request with the specific configuration of the component.

Note this is for the whole ConfigureScan request for a PST BEAM component, which includes checking both the beam and scan configuration is correct. This is due to the fact that clients of BEAM.MGMT only exposes a ConfigureScan request as it’s an Obs device.

Parameters

configuration (dict) – the configuration to validate.

class ska_pst.lmc.component.SubcomponentEventMessage(*, subcomponent_name: str, health_state: ska_control_model.HealthState, obs_state: ska_control_model.ObsState)

A data class used for passing events messages from subcomponents around.

When an event, such as health state change or going into a FAULT state, a subcomponent can create a message of this type and place it on the event queue that the BEAM component manager can listen to.

This class is designed to avoid the anti-pattern of callback hell. Rather than the BEAM component manager having to pass multiple callable functions to subcomponents or providing a reference to itself in the subcomponent managers, only a queue.Queue is needed to be shared and the BEAM component manger can handle the events with a background task.

health_state: ska_control_model.HealthState

The current health state of the subcomponent.

obs_state: ska_control_model.ObsState

The current observing state of the subcomponent.

subcomponent_name: str

The name of the subcomponent that created the event.

ska_pst.lmc.component.as_device_attribute_name(attr_name: str) → str

Convert attribute name to a TANGO device attribute name.

Device attribute names should be in lower camel case (i.e. availableDiskSpace) not Python’s snake case (i.e. available_disk_space). This is a utility method that makes the conversion easier.

Parameters

attr_name (str) – the attribute name to convert from snake case.

Returns

the lower camel case version of input string.

Return type

str

ska_pst.lmc.component.ascii_header_to_protobuf(ascii_header: ska_pydada.AsciiHeader) → ska_pst.grpc.lmc.ska_pst_lmc_pb2.AsciiHeaderProto

Convert an AsciiHeader instance to a protobuf equivalent message.

The protobuf AsciiHeader message uses the NewType pattern to wrap a string which allows for type safety on both the Python and C++ side of the LMC.

Parameters

ascii_header (AsciiHeader) – input ascii header that will be converted a protobuf message.

Returns

a protobuf message equivalent of the input AsciiHeader.

Return type

AsciiHeaderProto

ska_pst.lmc.component.protobuf_to_ascii_header(msg: ska_pst.grpc.lmc.ska_pst_lmc_pb2.AsciiHeaderProto) → ska_pydada.AsciiHeader

Convert a protobuf AsciiHeader message to a Python AsciiHeader.

Parameters

msg (AsciiHeaderProto) – input protobuf message.

Returns

the input message as an AsciiHeader instance.

Return type

AsciiHeader