health

README

Nurse healthcheck service

nurse leverages Go's pprof tooling to monitor various system and application metrics. The operator can configure per-resource thresholds which, when surpassed, trigger the gathering of a suite of metrics which are then dumped to disk along with a nurse.log file describing the reason for the dump.

To enable nurse, set the NURSE_TOML_PATH environment variable and create a nurse.toml file with the following schema:

profile-root           = "/tmp/nurse" # where the profiles will be dumped
poll-interval          = "1m"         # frequency with which to check thresholds
gather-duration        = "60s"        # how long to gather samples when profiling
max-profile-size       = "100mb"      # maximum size of the profile
cpu-profile-rate       = "5"          # samples per second
mem-profile-rate       = "5"          # samples per second
block-profile-rate     = "..."        # samples per second
mutex-profile-fraction = "..."        # fraction of mutex events that are reports (1 / rate)
mem-threshold          = "..."        # maximum memory usage of the process before a dump is triggered
goroutine-threshold    = "..."        # maximum number of goroutines before a dump is triggered

Documentation

Index

• type CheckFunc
• type Meta
• type Nurse
  • func NewNurse(cfg NurseConfig) *Nurse
  • func (n *Nurse) AddCheck(reason string, checkFunc CheckFunc)
  • func (n *Nurse) Close() error
  • func (n *Nurse) GatherVitals(reason string, meta Meta)
  • func (n *Nurse) Start() error
• type NurseConfig
  • func MustReadConfigTOML(path string) NurseConfig
  • func ReadConfigTOML(path string) (NurseConfig, error)

Types

type CheckFunc

type CheckFunc func() (unwell bool, meta Meta)

type Meta

type Meta map[string]any

type Nurse

type Nurse struct {
	log.Logger
	// contains filtered or unexported fields
}

func NewNurse

func NewNurse(cfg NurseConfig) *Nurse

func (*Nurse) AddCheck

func (n *Nurse) AddCheck(reason string, checkFunc CheckFunc)

func (*Nurse) Close

func (n *Nurse) Close() error

func (*Nurse) GatherVitals

func (n *Nurse) GatherVitals(reason string, meta Meta)

func (*Nurse) Start

func (n *Nurse) Start() error

type NurseConfig

type NurseConfig struct {
	ProfileRoot          string         `toml:"profile-root"`
	PollInterval         time.Duration  `toml:"poll-interval"`
	GatherDuration       time.Duration  `toml:"gather-duration"`
	MaxProfileSize       utils.ByteSize `toml:"max-profile-size"`
	CPUProfileRate       int            `toml:"cpu-profile-rate"`
	MemProfileRate       int            `toml:"mem-profile-rate"`
	BlockProfileRate     int            `toml:"block-profile-rate"`
	MutexProfileFraction int            `toml:"mutex-profile-fraction"`
	MemThreshold         utils.ByteSize `toml:"mem-threshold"`
	GoroutineThreshold   int            `toml:"goroutine-threshold"`

	Logger log.Logger `toml:"-"`
}

func MustReadConfigTOML

func MustReadConfigTOML(path string) NurseConfig

func ReadConfigTOML

func ReadConfigTOML(path string) (NurseConfig, error)