SDP Controller Device

Introduction

The SDP controller Tango device is designed to provide the overall control and monitoring of the SDP. Its attributes report on the overall state of the system.

The present implementation of the SDP controller device does the following:

performs the state transitions in response to commands
reports the version of the SDP and its components
reports the status of SDP components
reports the health state of the system based on the component status

Commands

The commands change the device state as described below, but at present they have no other effect on SDP.

Commands
Command	Argument type	Return type	Action
On	None	None	Set device state to ON
Standby	None	None	Set device state to STANDBY
Off	None	None	Set device state to OFF

Attributes

Attributes
Attribute	Type	Read/Write	Values	Description
versionId	String	Read	Semantic version	Controller device server version
sdpVersion	String (JSON)	Read	SDP version information	Versions of the SDP and its components
adminMode	Enum	Read/Write	Admin mode values	SDP admin mode
healthState	Enum	Read	Health state values	SDP health state
healthInfo	Sequence (spectrum) of strings	Read	Health info attribute	SDP health information
components	String (JSON)	Read	Components attribute	Reports information on SDP components
capacityBufferStorage	Int	Read	Capacity Buffer Storage Space attribute	Reports available capacity buffer storage space in MB

Events

The controller device generates both change events and archive events for the following attributes:

State
versionId
sdpVersion
adminMode
healthState
healthInfo
components
capacityBufferStorage

SDP version information

This attribute contains information about the versions of the SDP and its components, formatted as a JSON string. The following is an example of the version information as provided by the SDP Helm chart.

{
  "version": "2.3.0",
  "components": {
    "console": {
      "devicename": null,
      "image": "artefact.skao.int/ska-sdp-console",
      "version": "1.2.1"
    },
    "dsconfig": {
      "devicename": null,
      "image": "artefact.skao.int/ska-tango-images-tango-dsconfig",
      "version": "1.5.13"
    },
    "etcd": {
      "devicename": null,
      "image": "artefact.skao.int/ska-sdp-etcd",
      "version": "3.5.21"
    },
    "helmdeploy": {
      "devicename": null,
      "image": "artefact.skao.int/ska-sdp-helmdeploy",
      "version": "1.0.5"
    },
    "lmc-controller": {
      "devicename": "test-sdp/control/0",
      "image": "registry.gitlab.com/ska-telescope/sdp/ska-sdp-lmc/ska-sdp-lmc",
      "version": "1.3.0-dev.c23f79c7c"
    },
    "lmc-queueconnector-01": {
      "devicename": "test-sdp/queueconnector/01",
      "image": "artefact.skao.int/ska-sdp-lmc-queue-connector",
      "version": "5.6.2"
    },
    "lmc-subarray-01": {
      "devicename": "test-sdp/subarray/01",
      "image": "registry.gitlab.com/ska-telescope/sdp/ska-sdp-lmc/ska-sdp-lmc",
      "version": "1.3.0-dev.c23f79c7c"
    },
    "proccontrol": {
      "devicename": null,
      "image": "artefact.skao.int/ska-sdp-proccontrol",
      "version": "1.4.0"
    },
    "qa-api": {
      "devicename": null,
      "image": "artefact.skao.int/ska-sdp-qa-data-api",
      "version": "1.6.0"
    },
    "qa-display": {
      "devicename": null,
      "image": "artefact.skao.int/ska-sdp-qa-display",
      "version": "2.1.0"
    }
  },
  "dependencies": {}
}

Admin mode values

The admin mode is a writeable attribute which is used to indicate the intended use of the SDP.

Admin mode values
adminMode	Description
ONLINE (0)	SDP can be used for normal operations
OFFLINE (1)	SDP should not be used
ENGINEERING (2)	SDP can be used for engineering purposes such as testing, debugging or commissioning
NOT_FITTED (3)	Treated as OFFLINE
RESERVED (4)	Treated as OFFLINE

Some of these values are not applicable to a software-only system like the SDP, but they are implemented for compatibility with the standard interface.

If the admin mode is set to OFFLINE, NOT_FITTED or RESERVED the device state is set to DISABLE, but there are no other changes to the behaviour of the system.

Health state values

Health state values
healthState	Description
OK (0)	SDP is functioning as expected
DEGRADED (1)	SDP can only provide part of its functionality
FAILED (2)	SDP is unable to function
UNKNOWN (3)	Controller device is unable to determine the health of the SDP

The health state is based on the status of the SDP components:

if all components are online: health state is OK
if the Helm Deployer or the Processing Controller is offline: health state is DEGRADED
if any of the subarrays are offline: health state is DEGRADED
if all subarrays are offline: health state is FAILED

Health info attribute

Reports the reason for the Health state being DEGRADED or FAILED. It lists the SDP components that cannot be reached and describes what this means for the SDP system.

The components that affect this value are the Processing Controller, Helm Deployer, and the Subarray devices.

Components attribute

Reports various information on components in the form of a JSON string. Currently, it contains status information of the:

Subarray device(s)
Processing Controller
Helm Deployer

When queried, it checks that each component is online i.e. that it is reported as being “alive” in the configuration database.

Example string:

{
  "lmc-controller": {
    "status": "ONLINE"
  },
  "lmc-subarray-01": {
    "status": "ONLINE"
  },
  "proccontrol": {
    "status": "OFFLINE",
    "error": "SDP Processing Controller is OFFLINE: cannot start new processing blocks"
  },
  "helmdeploy": {
    "status": "ONLINE"
  }
}

Capacity Buffer Storage Space attribute

From the resource information stored in the configuration database (/resource of kind capacity-buffer-storage), the capacityBufferStorage attribute exposes the quantity (MB).

This value is the free storage space available on the volume and is calculated by the buffer manager with the following calculation: Free Space (quantity) = Total Volume Capacity - Orphaned Data - Contingency Reserve

When the resource information is unavailable, the capacityBufferStorage attribute defaults to -1.