infiniband

Documentation

Index

• Variables
• func CheckInfiniband(ctx context.Context, ibstatCommand string, threshold ExpectedPortStates) error
• func ValidateIbstatOutput(s string) error
• type ExpectedPortStates
  • func SupportsInfinibandPortRate(gpuProductName string) (ExpectedPortStates, error)
• type IBStatCard
• type IBStatCards
  • func ParseIBStat(input string) (IBStatCards, error)
  • func (cards IBStatCards) CheckPortsAndRate(atLeastPorts int, atLeastRate int) error
  • func (cards IBStatCards) Match(expectedPhysicalState string, expectedState string, atLeastRate int) []string
• type IBStatPort
• type IbstatOutput
  • func GetIbstatOutput(ctx context.Context, ibstatCommands []string) (*IbstatOutput, error)

Variables

var (
	ErrIbstatOutputBrokenStateDown        = errors.New("ibstat output unexpected; found State: Down (check the physical switch)")
	ErrIbstatOutputBrokenPhysicalDisabled = errors.New("ibstat output unexpected; found Physical state: Disabled (check the physical switch)")
)

var (
	ErrIbstatOutputEmpty       = errors.New("ibstat returned empty output")
	ErrIbstatOutputNoCardFound = errors.New("parsed ibstat output does not contain any card")
)

var ErrNoExpectedPortStates = errors.New("no expected port states found (not supported)")

var ErrNoIbstatCommand = errors.New("ibstat not found. cannot check ib state")

Functions

func CheckInfiniband

func CheckInfiniband(ctx context.Context, ibstatCommand string, threshold ExpectedPortStates) error

CheckInfiniband checks if the infiniband ports are up and running with the expected thresholds.

func ValidateIbstatOutput

func ValidateIbstatOutput(s string) error

Types

type ExpectedPortStates

type ExpectedPortStates struct {
	// The minimum number of ports.
	// If not set, it defaults to 0.
	AtLeastPorts int `json:"at_least_ports"`

	// The expected rate in Gb/sec.
	// If not set, it defaults to 0.
	AtLeastRate int `json:"at_least_rate"`
}

Configures the expected state of the ports.

func SupportsInfinibandPortRate

func SupportsInfinibandPortRate(gpuProductName string) (ExpectedPortStates, error)

type IBStatCard

type IBStatCard struct {
	Name            string     `json:"CA name"`
	Type            string     `json:"CA type"`
	NumPorts        string     `json:"Number of ports"`
	FirmwareVersion string     `json:"Firmware version"`
	HardwareVersion string     `json:"Hardware version"`
	NodeGUID        string     `json:"Node GUID"`
	SystemImageGUID string     `json:"System image GUID"`
	Port1           IBStatPort `json:"Port 1"`
}

type IBStatCards

type IBStatCards []IBStatCard

func ParseIBStat

func ParseIBStat(input string) (IBStatCards, error)

ParseIBStat parses ibstat output and returns YAML representation. Returns ErrIbstatOutputEmpty if the input is empty.

func (IBStatCards) CheckPortsAndRate

func (cards IBStatCards) CheckPortsAndRate(atLeastPorts int, atLeastRate int) error

CheckPortsAndRate checks if the number of active IB ports matches expectations

func (IBStatCards) Match

func (cards IBStatCards) Match(expectedPhysicalState string, expectedState string, atLeastRate int) []string

Match returns the IB port names whose physical state, state, and "Port 1"."Rate" match the expected values. The specified rate is the threshold for "Port 1"."Rate", where it evaluates with ">=" operator (e.g., count all the cards whose rate is >= 400).

If the `expectedPhysicalState` is empty, it matches all states. If the `expectedState` is empty, it matches all states.

type IBStatPort

type IBStatPort struct {
	State         string `json:"State"`
	PhysicalState string `json:"Physical state"`
	Rate          int    `json:"Rate"`
	BaseLid       int    `json:"Base lid"`
	LinkLayer     string `json:"Link layer"`
}

type IbstatOutput

type IbstatOutput struct {
	Parsed IBStatCards `json:"parsed,omitempty"`
	Raw    string      `json:"raw"`
	Errors []string    `json:"errors,omitempty"`
}

func GetIbstatOutput

func GetIbstatOutput(ctx context.Context, ibstatCommands []string) (*IbstatOutput, error)