Processor Resources

mindx-dl-deviceinfo-<nodename>

Table 1 lists the NPU information reported by Ascend Device Plugin.

**Table 1** DeviceInfoCfg
Parameter	Meaning	Remarks
huawei.com/Ascend910	Name of the available processor on the current node. If there are multiple processors, use commas (,) to separate them.	This field has been unavailable and will not be displayed in later versions. By default, the available processors of a node are maintained by Volcano, and this field does not take effect. To make the field take effect, change the value of the Volcano configuration parameter self-maintain-available-card to false.
huawei.com/Ascend910-NetworkUnhealthy	Name of the processor with unhealthy network on the current node. If there are multiple processors, use commas (,) to separate them.	-
huawei.com/Ascend910-Unhealthy	Name of the unhealthy processor on the current node. If there are multiple processors, use commas (,) to separate them.	-
huawei.com/Ascend910-Recovering	Processor being restored on the current node. If there are multiple processors, use commas (,) to separate them.	-
huawei.com/Ascend910-Fault	Detailed fault information about a processor	Array object, including fault_type, npu_name, large_model_fault_level, fault_level, fault_handling, fault_code, and fault_time_and_level_map.
-fault_type	Fault type	CardUnhealthy: processor fault. CardNetworkUnhealthy: processor network fault. NodeUnhealthy: node fault.
-npu_name	Name of the faulty processor. This parameter is left empty if the node is faulty.	String
-large_model_fault_level	Fault handling type. This parameter is left empty if the node is faulty.	NotHandleFault: requires no handling. RestartRequest: re-executes inference requests in the inference scenario, or re-executes training services in the training scenario. RestartBusiness: re-executes services. FreeRestartNPU: resets an idle processor when faults affect service execution. RestartNPU: directly resets processors and re-executes services. SeparateNPU: isolates processors. PreSeparateNPU: pre-isolates processors and determines whether to perform rescheduling based on the actual running status of the training job. NOTE: The functions of large_model_fault_level, fault_level and fault_handling are the same. fault_handling is recommended. When an inference job subscribes to fault information, a RestartRequest fault on its inference card will not trigger rescheduling if the fault lasts 60 seconds or less. If the fault persists for more than 60 seconds, the related processor is isolated and the job is rescheduled.
-fault_level
-fault_handling
-fault_code	Fault code, a string of characters separated by commas (,).	For details about processor fault codes, see Processor Fault Code Reference Documents.
-fault_time_and_level_map	Fault code, fault occurrence time, and fault handling level.	-
SuperPodID	SuperPoD ID	String
ServerIndex	Relative position of the current node in a SuperPoD	When SuperPodID or ServerIndex reported by the driver is 0xffffffff, the corresponding value of SuperPodID or ServerIndex is -1. The value of SuperPodID or ServerIndex is -2 in the following situations: The current device does not support the query of SuperPoD information. The SuperPoD information fails to be obtained due to a driver problem.
CheckCode	Verification code	-

Table 2 describes the interconnect device fault information reported by Ascend Device Plugin.

**Table 2** SwitchInfoCfg
Resource	Meaning	Description
FaultCode	List of interconnect device fault codes on the current node	Array object, including the EventType, AssembledFaultCode, PeerPortDevice, PeerPortId, SwitchChipId, SwitchPortId, Severity, Assertion, and AlarmRaisedTime fields.
-EventType	Alarm ID	-
-AssembledFaultCode	Fault code	-
-PeerPortDevice	Type of the device interconnected	0: CPU 1: NPU 2: SW 0xFFFF: N/A
-PeerPortId	ID of the interconnected device	-
-SwitchChipId	ID of the faulty UnifiedBus chip	Numbered from 0
-SwitchPortId	ID of the faulty UnifiedBus port	Numbered from 0
-Severity	Fault severity	0: warning 1: minor 2: major 3: critical
-Assertion	Event type	0: fault recovered 1: fault occurred 2: notification event
-AlarmRaisedTime	Time when a fault or event occurs	-
FaultLevel	Fault handling level of the current node	The highest fault level in FaultCode is used. The options are NotHandle, SubHealthFault, Separate, and RestartRequest.
UpdateTime	Update time when a fault is reported	-
NodeStatus	Health status of the current node	Corresponds to the value of FaultLevel, including NotHandle: Healthy, SubHealthFault: SubHealthy, Separate: UnHealthy, and RestartRequest: UnHealthy.
FaultTimeAndLevelMap	List of fault occurrence time and fault handling levels	Array object, including fault code, faulty UnifiedBus chip ID, faulty UnifiedBus port ID, fault_time, and fault_level. The key consists of the fault code, faulty UnifiedBus chip ID, and faulty UnifiedBus port ID, which are connected by underscores (_).
-fault_time	Time when a fault occurs	-
-fault_level	Fault handling level	-

Table 3 describes the description information in the ConfigMap of Ascend Device Plugin.

**Table 3** Description
Resource	Meaning	Description
Description	Description	The available processor information of the node in this ConfigMap has been unavailable. By default, the available processors of a node are maintained by Volcano, and the processor information maintained in this ConfigMap does not take effect. To make the field take effect, change the value of the Volcano configuration parameter self-maintain-available-card to false.

Table 4 describes the NPU fault information reported by Ascend Device Plugin. The object is named <device-plugin-pod-name>.<reporting time><faulty processor ID>, and its type is Event.

The following table describes only the fields related to MindCluster services. For details about more fields, see Event core.

**Table 4** NPU fault information
Resource	Meaning	Description
type	Event severity	It can only be Warning.
message	Event content, including the node name, processor ID, fault occurrence or recovery type, fault code, and fault severity.	String
reason	Reason for reporting an event	Recovery: fault recovered Occur: fault occurred Notice: one-off fault notification
action	Fault level	String. For details, see Table 1.
source	Fault source	Structure, specifies the node where a fault occurs.
eventTime	Time when a fault occurs	Timestamp
involvedObject	Object involved in a fault	Structure. The Kind, Namespace, and Name fields specify the pod name of Ascend Device Plugin. Once specified, the event can be queried directly through the Event object or retrieved when viewing the current pod details.
reportingComponent	Event controller	The value can only be device-plugin.
reportingInstance	Event reporting instance	String. The value is the Ascend Device Plugin pod name.

deviceNameCustomization.json

Device names can be customized in deviceNameCustomization.json. When compiling the Ascend Device Plugin image, place this file in the same directory as the binary package to change the resource type and resource name displayed by Ascend Device Plugin to custom names.

**Table 5** deviceNameCustomization.json description
Resource	Description	Value
ResourceType	Initial device name, which is mandatory.	Only Ascend 910, Ascend 310, and Ascend 310P are supported.
DevicePublicType	Type of the device displayed externally, for example, huawei.com/Ascend910. This parameter is mandatory.	Only the xxx.xxx/xxx format is supported. The value of xxx can contain 10 to 32 characters, including uppercase letters, lowercase letters, and digits.
DevicePublicNamePre	Prefix of the device name displayed externally, for example, Ascend910-. For the name of the device displayed externally, Ascend Device Plugin appends the physical ID of the processor to the prefix. This parameter is mandatory.	The value can contain 2 to 16 characters, including uppercase letters, lowercase letters, hyphens (-), and digits. The value must start with an uppercase or lowercase letter.
PodConfigurationName	Details of the mounted processor information displayed in the pod annotation. This parameter is mandatory when ResourceType is set to Ascend910.	The value can contain 10 to 63 characters, including uppercase letters, lowercase letters, hyphens (-), slashes (/), periods (.), and digits. The value must start with an uppercase or lowercase letter and end with an uppercase or lowercase letter or digit.

Parent topic: Ascend Device Plugin