common

Documentation

Overview

Package common contains common types and functions used across multiple components.

Index

• type EventType
  • func EventTypeFromString(s string) EventType
• type RepairActionType
• type SuggestedActions
  • func (s *SuggestedActions) Add(other *SuggestedActions)
  • func (s *SuggestedActions) RequiresCheckUserAppAndGPU() bool
  • func (s *SuggestedActions) RequiresReboot() bool
  • func (s *SuggestedActions) RequiresRepair() bool

Types

type EventType

type EventType string

const (
	EventTypeUnknown EventType = "Unknown"

	// EventTypeInfo represents a general event that requires no action.
	// Info - Informative, no further action needed.
	EventTypeInfo EventType = "Info"

	// EventTypeWarning represents an event that may impact workloads.
	// Warning - Some issue happened but no further action needed, expecting automatic recovery.
	EventTypeWarning EventType = "Warning"

	// EventTypeCritical represents an event that is definitely impacting workloads
	// and requires immediate attention.
	// Critical - Some critical issue happened thus action required, not a hardware issue.
	EventTypeCritical EventType = "Critical"

	// EventTypeFatal represents a fatal event that impacts wide systems
	// and requires immediate attention and action.
	// Fatal - Fatal/hardware issue occurred thus immediate action required, may require reboot/hardware repair.
	EventTypeFatal EventType = "Fatal"
)

func EventTypeFromString

func EventTypeFromString(s string) EventType

type RepairActionType

type RepairActionType string

const (
	// RepairActionTypeIgnoreNoActionRequired represents a suggested action to ignore the issue,
	// meaning no action is needed until further notice.
	RepairActionTypeIgnoreNoActionRequired RepairActionType = "IGNORE_NO_ACTION_REQUIRED"

	// RepairActionTypeRebootSystem represents a suggested action to reboot the system.
	// Specific to NVIDIA GPUs, this implies GPU reset by rebooting the system.
	RepairActionTypeRebootSystem RepairActionType = "REBOOT_SYSTEM"

	// RepairActionTypeHardwareInspection represents a suggested action for hardware inspection
	// and repair if any issue is found. This often involves data center (or cloud provider) support
	// to physically check/repair the machine.
	RepairActionTypeHardwareInspection RepairActionType = "HARDWARE_INSPECTION"

	// RepairActionTypeCheckUserApp represents a suggested action to check the user application.
	// For instance, NVIDIA may report XID 45 as user app error, but the underlying GPU might have other issues
	// thus requires further diagnosis of the application and the GPU.
	RepairActionTypeCheckUserAppAndGPU RepairActionType = "CHECK_USER_APP_AND_GPU"
)

type SuggestedActions

type SuggestedActions struct {
	// References to the descriptions.
	References []string `json:"references,omitempty"`

	// A list of reasons and descriptions for the suggested actions.
	Descriptions []string `json:"descriptions"`

	// A list of repair actions to mitigate the issue.
	RepairActions []RepairActionType `json:"repair_actions"`
}

SuggestedActions represents a set of suggested actions to mitigate an issue.

func (*SuggestedActions) Add

func (s *SuggestedActions) Add(other *SuggestedActions)

func (*SuggestedActions) RequiresCheckUserAppAndGPU

func (s *SuggestedActions) RequiresCheckUserAppAndGPU() bool

func (*SuggestedActions) RequiresReboot

func (s *SuggestedActions) RequiresReboot() bool

func (*SuggestedActions) RequiresRepair

func (s *SuggestedActions) RequiresRepair() bool