@balena/jellyfish-metrics NPM

Jellyfish Metrics

This library gathers Prometheus metrics and exposes them to be scraped by balena-monitor.

Usage

Below is an example how to use this library:

const metrics = require('@balena/jellyfish-metrics')

// Start server to expose gathered metrics data.
metrics.startServer(context, portNumber)

// Mark that a card was read from the database.
metrics.markCardReadFromDatabase(card)

Documentation

Adding metrics

A module that gathers and exposes Prometheus metrics.

metrics
- static
- inner
  - ~measureAsync(name, labels, fn) ⇒ Any
  - ~isCard(card) ⇒ Boolean
  - ~getAsyncMeasureFn(prefix) ⇒ Any

metrics.actorFromContext(context) ⇒ String

Kind: static method of metrics
Summary: Extract actor name from context ID
Returns: String - actor name

Param	Type	Description
context	Object	caller context

Example

const actorName = exports.actorFromContext(context)

metrics.initExpress(context) ⇒ Object

Kind: static method of metrics
Summary: Create express app using metrics and expose data on /metrics
Returns: Object - express app

Param	Type	Description
context	Object	execution context

Example

const application = metrics.initExpress(context)

metrics.startServer(context, port)

Expose gathered metrics on /metrics Reassign port to random port number on collision

Kind: static method of metrics

Param	Type	Description
context	Object	execution context
port	Number	port to expose metrics on

Example

metrics.startServer(context, 9000)

metrics.markCardInsert(card)

Kind: static method of metrics
Summary: Mark that a card was inserted

Param	Type	Description
card	Object	card that was inserted

Example

metrics.markCardInsert(card)

metrics.markCardUpsert(card)

Kind: static method of metrics
Summary: Mark that a card was upserted

Param	Type	Description
card	Object	card that was upserted

Example

metrics.markCardUpsert(card)

metrics.markCardReadFromDatabase(card)

Kind: static method of metrics
Summary: Mark that a card was read from the database

Param	Type	Description
card	Object	card that was read the database

Example

metrics.markCardReadFromDatabase(card)

metrics.markCardReadFromCache(card)

Kind: static method of metrics
Summary: Mark that a card was read from cache

Param	Type	Description
card	Object	card that was read from cache

Example

metrics.markCardReadFromCache(card)

metrics.markBackSync(integration)

Kind: static method of metrics
Summary: Mark that a card has been created due to back-sync

Param	Type	Description
integration	String	name of integration

Example

metrics.markBackSync('front')

metrics.markActionRequest(action)

Kind: static method of metrics
Summary: Mark that an action request was received

Param	Type	Description
action	String	action name

Example

metrics.markActionRequest('action-create-card')

metrics.markQueueConcurrency()

Kind: static method of metrics
Summary: Expose current queue concurrency setting
Example

metrics.markQueueConcurrency()

metrics.markJobAdd(action, id)

Kind: static method of metrics
Summary: Mark that a new job was added to the queue

Param	Type	Description
action	String	action name
id	String	id of the worker

Example

metrics.markJobAdd('action-create-card', context.id)

metrics.markJobDone(action, id, timestamp)

Kind: static method of metrics
Summary: Mark that a job in the queue has completed

Param	Type	Description
action	String	action name
id	String	id of the worker
timestamp	String	when action was completed

Example

const action = 'action-create-card'
const timestamp = '2020-06-08T09:33:27.481Z'
metrics.markJobDone(action, context.id, timestamp)

metrics.measureMirror(integration, fn) ⇒ Any

Kind: static method of metrics
Summary: Execute a mirror, marking duration and totals
Returns: Any - mirror result

Param	Type	Description
integration	Object	name of external integration
fn	Promise	mirror function to execute

Example

const result = await metrics.measureMirror('github', mirror())

metrics.measureTranslate(integration, fn) ⇒ Any

Kind: static method of metrics
Summary: Execute a translate, marking duration and totals
Returns: Any - translate result

Param	Type	Description
integration	Object	name of external integration
fn	Promise	mirror function to execute

Example

const result = await metrics.measureTranslate('github', translate())

metrics.measureHttpQuery(fn) ⇒ Any

Kind: static method of metrics
Summary: Measure the duration of a request to the /query api endpoint
Returns: Any - api result

Param	Type	Description
fn	Promise	api function to execute

metrics.measureHttpType(fn) ⇒ Any

Kind: static method of metrics
Summary: Measure the duration of a request to the /type api endpoint
Returns: Any - api result

Param	Type	Description
fn	Promise	api function to execute

metrics.measureHttpId(fn) ⇒ Any

Kind: static method of metrics
Summary: Measure the duration of a request to the /id api endpoint
Returns: Any - api result

Param	Type	Description
fn	Promise	api function to execute

metrics.measureHttpSlug(fn) ⇒ Any

Kind: static method of metrics
Summary: Measure the duration of a request to the /slug api endpoint
Returns: Any - api result

Param	Type	Description
fn	Promise	api function to execute

metrics.measureHttpAction(fn) ⇒ Any

Kind: static method of metrics
Summary: Measure the duration of a request to the /action api endpoint
Returns: Any - api result

Param	Type	Description
fn	Promise	api function to execute

metrics.measureHttpWhoami(fn) ⇒ Any

Kind: static method of metrics
Summary: Measure the duration of a request to the /action api endpoint
Returns: Any - api result

Param	Type	Description
fn	Promise	api function to execute

metrics.markSqlGenTime(ms)

Kind: static method of metrics
Summary: Mark how long it took to generate an SQL query from a JSON schema

Param	Type	Description
ms	Number	number of milliseconds it took to generate the query

metrics.markQueryTime(ms)

Kind: static method of metrics
Summary: Mark how long it took to execute an SQL query

Param	Type	Description
ms	Number	number of milliseconds it took to execute the query

metrics.markStreamOpened(context, table)

Kind: static method of metrics
Summary: Mark that a new stream was opened

Param	Type	Description
context	Object	caller context
table	String	table name

Example

metrics.markStreamOpened(context, 'cards')

metrics.markStreamClosed(context, table)

Kind: static method of metrics
Summary: Mark that a stream was closed

Param	Type	Description
context	Object	caller context
table	String	table name

Example

metrics.markStreamClosed(context, 'cards')

metrics.markStreamLinkQuery(context, table, change)

Kind: static method of metrics
Summary: Mark that a stream is querying links

Param	Type	Description
context	Object	caller context
table	String	table name
change	Object	change event object

Example

metrics.markStreamLinkQuery(context, 'cards', change)

metrics.markStreamError(context, table)

Kind: static method of metrics
Summary: Mark that a stream error has occurred

Param	Type	Description
context	Object	caller context
table	String	table name

Example

metrics.markStreamError()

metrics~measureAsync(name, labels, fn) ⇒ Any

Kind: inner method of metrics
Summary: Measure duration of a promise execution and add to metrics
Returns: Any - promise execution result

Param	Type	Description
name	String	metric name
labels	Object \| Undefined	metric labels
fn	Promise	function to execute and measure

Example

const result = await measureAsync('my_metric', { ... }, myFunction, ...params)

metrics~isCard(card) ⇒ Boolean

Kind: inner method of metrics
Summary: Checks if an object looks to be a valid card or not.
Returns: Boolean - validation result, true if card, false if not

Param	Type	Description
card	Object	object to validate

Example

const result = isCard(card)

metrics~getAsyncMeasureFn(prefix) ⇒ Any

Kind: inner method of metrics
Summary: Generates a generic measurement wrapper for an async function, that tracks total calls, total failures and duration
Returns: Any - api result