luxolus
/
open-nomad
2
0
Fork 0
Code Issues 1 Pull Requests Packages Releases Wiki Activity

Merge pull request #3162 from hashicorp/f-tagged-metrics-api 

Tagged metrics API
This commit is contained in:
Chelsea Komlo 2017-09-07 18:03:15 -04:00 committed by GitHub
parent dbab588d60 e56da12857
commit 187c1568aa
16 changed files with 477 additions and 62 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
command/agent/agent.go
View File

@ -13,6 +13,7 @@ import (
"sync/atomic"
"time"
metrics "github.com/armon/go-metrics"
"github.com/hashicorp/consul/api"
version "github.com/hashicorp/go-version"
"github.com/hashicorp/nomad/client"
@ -66,15 +67,18 @@ type Agent struct {
shutdown bool
shutdownCh chan struct{}
shutdownLock sync.Mutex
InmemSink *metrics.InmemSink
}
// NewAgent is used to create a new agent with the given configuration
func NewAgent(config *Config, logOutput io.Writer) (*Agent, error) {
func NewAgent(config *Config, logOutput io.Writer, inmem *metrics.InmemSink) (*Agent, error) {
a := &Agent{
config: config,
logger: log.New(logOutput, "", log.LstdFlags|log.Lmicroseconds),
logOutput: logOutput,
shutdownCh: make(chan struct{}),
InmemSink: inmem,
}
if err := a.setupConsul(config.Consul); err != nil {
@ -331,9 +335,13 @@ func (a *Agent) clientConfig() (*clientconfig.Config, error) {
conf.ConsulConfig = a.config.Consul
conf.VaultConfig = a.config.Vault
// Set up Telemetry configuration
conf.StatsCollectionInterval = a.config.Telemetry.collectionInterval
conf.PublishNodeMetrics = a.config.Telemetry.PublishNodeMetrics
conf.PublishAllocationMetrics = a.config.Telemetry.PublishAllocationMetrics
conf.DisableTaggedMetrics = a.config.Telemetry.DisableTaggedMetrics
conf.BackwardsCompatibleMetrics = a.config.Telemetry.BackwardsCompatibleMetrics
// Set the TLS related configs
conf.TLSConfig = a.config.TLSConfig
@ -489,7 +497,6 @@ func (a *Agent) setupClient() error {
}
}
// Create the client
client, err := client.NewClient(conf, a.consulCatalog, a.consulService, a.logger)
if err != nil {
return fmt.Errorf("client setup failed: %v", err)

24
command/agent/agent_test.go
View File

@ -12,6 +12,7 @@ import (
"github.com/hashicorp/nomad/helper"
sconfig "github.com/hashicorp/nomad/nomad/structs/config"
"github.com/stretchr/testify/assert"
)
func getPort() int {
@ -316,6 +317,29 @@ func TestAgent_ClientConfig(t *testing.T) {
}
}
// Clients should inherit telemetry configuration
func TestAget_Client_TelemetryConfiguration(t *testing.T) {
assert := assert.New(t)
conf := DefaultConfig()
conf.DevMode = true
conf.Telemetry.DisableTaggedMetrics = true
conf.Telemetry.BackwardsCompatibleMetrics = true
a := &Agent{config: conf}
c, err := a.clientConfig()
assert.Nil(err)
telemetry := conf.Telemetry
assert.Equal(c.StatsCollectionInterval, telemetry.collectionInterval)
assert.Equal(c.PublishNodeMetrics, telemetry.PublishNodeMetrics)
assert.Equal(c.PublishAllocationMetrics, telemetry.PublishAllocationMetrics)
assert.Equal(c.DisableTaggedMetrics, telemetry.DisableTaggedMetrics)
assert.Equal(c.BackwardsCompatibleMetrics, telemetry.BackwardsCompatibleMetrics)
}
// TestAgent_HTTPCheck asserts Agent.agentHTTPCheck properly alters the HTTP
// API health check depending on configuration.
func TestAgent_HTTPCheck(t *testing.T) {

27
command/agent/command.go
View File

@ -15,12 +15,12 @@ import (
"syscall"
"time"
"github.com/armon/go-metrics"
metrics "github.com/armon/go-metrics"
"github.com/armon/go-metrics/circonus"
"github.com/armon/go-metrics/datadog"
"github.com/hashicorp/consul/lib"
"github.com/hashicorp/go-checkpoint"
"github.com/hashicorp/go-syslog"
checkpoint "github.com/hashicorp/go-checkpoint"
gsyslog "github.com/hashicorp/go-syslog"
"github.com/hashicorp/logutils"
"github.com/hashicorp/nomad/helper/flag-helpers"
"github.com/hashicorp/nomad/helper/gated-writer"
@ -331,9 +331,9 @@ func (c *Command) setupLoggers(config *Config) (*gatedwriter.Writer, *logWriter,
}
// setupAgent is used to start the agent and various interfaces
func (c *Command) setupAgent(config *Config, logOutput io.Writer) error {
func (c *Command) setupAgent(config *Config, logOutput io.Writer, inmem *metrics.InmemSink) error {
c.Ui.Output("Starting Nomad agent...")
agent, err := NewAgent(config, logOutput)
agent, err := NewAgent(config, logOutput, inmem)
if err != nil {
c.Ui.Error(fmt.Sprintf("Error starting agent: %s", err))
return err
@ -444,13 +444,14 @@ func (c *Command) Run(args []string) int {
}
// Initialize the telemetry
if err := c.setupTelemetry(config); err != nil {
inmem, err := c.setupTelemetry(config)
if err != nil {
c.Ui.Error(fmt.Sprintf("Error initializing telemetry: %s", err))
return 1
}
// Create the agent
if err := c.setupAgent(config, logOutput); err != nil {
if err := c.setupAgent(config, logOutput, inmem); err != nil {
logGate.Flush()
return 1
}
@ -619,7 +620,7 @@ func (c *Command) handleReload(config *Config) *Config {
}
// setupTelemetry is used ot setup the telemetry sub-systems
func (c *Command) setupTelemetry(config *Config) error {
func (c *Command) setupTelemetry(config *Config) (*metrics.InmemSink, error) {
/* Setup telemetry
Aggregate on 10 second intervals for 1 minute. Expose the
metrics over stderr when there is a SIGUSR1 received.
@ -646,7 +647,7 @@ func (c *Command) setupTelemetry(config *Config) error {
if telConfig.StatsiteAddr != "" {
sink, err := metrics.NewStatsiteSink(telConfig.StatsiteAddr)
if err != nil {
return err
return inm, err
}
fanout = append(fanout, sink)
}
@ -655,7 +656,7 @@ func (c *Command) setupTelemetry(config *Config) error {
if telConfig.StatsdAddr != "" {
sink, err := metrics.NewStatsdSink(telConfig.StatsdAddr)
if err != nil {
return err
return inm, err
}
fanout = append(fanout, sink)
}
@ -664,7 +665,7 @@ func (c *Command) setupTelemetry(config *Config) error {
if telConfig.DataDogAddr != "" {
sink, err := datadog.NewDogStatsdSink(telConfig.DataDogAddr, config.NodeName)
if err != nil {
return err
return inm, err
}
fanout = append(fanout, sink)
}
@ -700,7 +701,7 @@ func (c *Command) setupTelemetry(config *Config) error {
sink, err := circonus.NewCirconusSink(cfg)
if err != nil {
return err
return inm, err
}
sink.Start()
fanout = append(fanout, sink)
@ -714,7 +715,7 @@ func (c *Command) setupTelemetry(config *Config) error {
metricsConf.EnableHostname = false
metrics.NewGlobal(metricsConf, inm)
}
return nil
return inm, nil
}
// setupSCADA is used to start a new SCADA provider and listener,

4
command/agent/config-test-fixtures/basic.hcl
View File

@ -60,8 +60,6 @@ client {
gc_inode_usage_threshold = 91
gc_max_allocs = 50
no_host_uuid = false
disable_tagged_metrics = true
backwards_compatible_metrics = true
}
server {
enabled = true
@ -98,6 +96,8 @@ telemetry {
collection_interval = "3s"
publish_allocation_metrics = true
publish_node_metrics = true
disable_tagged_metrics = true
backwards_compatible_metrics = true
}
leave_on_interrupt = true
leave_on_terminate = true

33
command/agent/config.go
View File

@ -229,14 +229,6 @@ type ClientConfig struct {
// NoHostUUID disables using the host's UUID and will force generation of a
// random UUID.
NoHostUUID *bool `mapstructure:"no_host_uuid"`
// DisableTaggedMetrics disables a new version of generating metrics which
// uses tags
DisableTaggedMetrics bool `mapstructure:"disable_tagged_metrics"`
// BackwardsCompatibleMetrics allows for generating metrics in a simple
// key/value structure as done in older versions of Nomad
BackwardsCompatibleMetrics bool `mapstructure:"backwards_compatible_metrics"`
}
// ACLConfig is configuration specific to the ACL system
@ -371,6 +363,14 @@ type Telemetry struct {
PublishAllocationMetrics bool `mapstructure:"publish_allocation_metrics"`
PublishNodeMetrics bool `mapstructure:"publish_node_metrics"`
// DisableTaggedMetrics disables a new version of generating metrics which
// uses tags
DisableTaggedMetrics bool `mapstructure:"disable_tagged_metrics"`
// BackwardsCompatibleMetrics allows for generating metrics in a simple
// key/value structure as done in older versions of Nomad
BackwardsCompatibleMetrics bool `mapstructure:"backwards_compatible_metrics"`
// Circonus: see https://github.com/circonus-labs/circonus-gometrics
// for more details on the various configuration options.
// Valid configuration combinations:
@ -1105,14 +1105,6 @@ func (a *ClientConfig) Merge(b *ClientConfig) *ClientConfig {
result.NoHostUUID = b.NoHostUUID
}
if b.DisableTaggedMetrics {
result.DisableTaggedMetrics = b.DisableTaggedMetrics
}
if b.BackwardsCompatibleMetrics {
result.BackwardsCompatibleMetrics = b.BackwardsCompatibleMetrics
}
// Add the servers
result.Servers = append(result.Servers, b.Servers...)
@ -1214,6 +1206,15 @@ func (a *Telemetry) Merge(b *Telemetry) *Telemetry {
if b.CirconusBrokerSelectTag != "" {
result.CirconusBrokerSelectTag = b.CirconusBrokerSelectTag
}
if b.DisableTaggedMetrics {
result.DisableTaggedMetrics = b.DisableTaggedMetrics
}
if b.BackwardsCompatibleMetrics {
result.BackwardsCompatibleMetrics = b.BackwardsCompatibleMetrics
}
return &result
}

4
command/agent/config_parse.go
View File

@ -357,8 +357,6 @@ func parseClient(result **ClientConfig, list *ast.ObjectList) error {
"gc_parallel_destroys",
"gc_max_allocs",
"no_host_uuid",
"disable_tagged_metrics",
"backwards_compatible_metrics",
}
if err := checkHCLKeys(listVal, valid); err != nil {
return err
@ -635,6 +633,8 @@ func parseTelemetry(result **Telemetry, list *ast.ObjectList) error {
"circonus_check_tags",
"circonus_broker_id",
"circonus_broker_select_tag",
"disable_tagged_metrics",
"backwards_compatible_metrics",
}
if err := checkHCLKeys(listVal, valid); err != nil {
return err

32
command/agent/config_parse_test.go
View File

@ -75,14 +75,12 @@ func TestConfig_Parse(t *testing.T) {
ReservedPorts: "1,100,10-12",
ParsedReservedPorts: []int{1, 10, 11, 12, 100},
},
GCInterval: 6 * time.Second,
GCParallelDestroys: 6,
GCDiskUsageThreshold: 82,
GCInodeUsageThreshold: 91,
GCMaxAllocs: 50,
NoHostUUID: helper.BoolToPtr(false),
DisableTaggedMetrics: true,
BackwardsCompatibleMetrics: true,
GCInterval: 6 * time.Second,
GCParallelDestroys: 6,
GCDiskUsageThreshold: 82,
GCInodeUsageThreshold: 91,
GCMaxAllocs: 50,
NoHostUUID: helper.BoolToPtr(false),
},
Server: &ServerConfig{
Enabled: true,
@ -113,14 +111,16 @@ func TestConfig_Parse(t *testing.T) {
ReplicationToken: "foobar",
},
Telemetry: &Telemetry{
StatsiteAddr: "127.0.0.1:1234",
StatsdAddr: "127.0.0.1:2345",
DisableHostname: true,
UseNodeName: false,
CollectionInterval: "3s",
collectionInterval: 3 * time.Second,
PublishAllocationMetrics: true,
PublishNodeMetrics: true,
StatsiteAddr: "127.0.0.1:1234",
StatsdAddr: "127.0.0.1:2345",
DisableHostname: true,
UseNodeName: false,
CollectionInterval: "3s",
collectionInterval: 3 * time.Second,
PublishAllocationMetrics: true,
PublishNodeMetrics: true,
DisableTaggedMetrics: true,
BackwardsCompatibleMetrics: true,
},
LeaveOnInt: true,
LeaveOnTerm: true,

16
command/agent/config_test.go
View File

@ -55,6 +55,8 @@ func TestConfig_Merge(t *testing.T) {
StatsdAddr: "127.0.0.1:8125",
DataDogAddr: "127.0.0.1:8125",
DisableHostname: false,
DisableTaggedMetrics: true,
BackwardsCompatibleMetrics: true,
CirconusAPIToken: "0",
CirconusAPIApp: "nomadic",
CirconusAPIURL: "http://api.circonus.com/v2",
@ -89,8 +91,6 @@ func TestConfig_Merge(t *testing.T) {
ReservedPorts: "1,10-30,55",
ParsedReservedPorts: []int{1, 2, 4},
},
DisableTaggedMetrics: true,
BackwardsCompatibleMetrics: true,
},
Server: &ServerConfig{
Enabled: false,
@ -185,6 +185,8 @@ func TestConfig_Merge(t *testing.T) {
DisableHostname: true,
PublishNodeMetrics: true,
PublishAllocationMetrics: true,
DisableTaggedMetrics: true,
BackwardsCompatibleMetrics: true,
CirconusAPIToken: "1",
CirconusAPIApp: "nomad",
CirconusAPIURL: "https://api.circonus.com/v2",
@ -226,12 +228,10 @@ func TestConfig_Merge(t *testing.T) {
ReservedPorts: "2,10-30,55",
ParsedReservedPorts: []int{1, 2, 3},
},
GCInterval: 6 * time.Second,
GCParallelDestroys: 6,
GCDiskUsageThreshold: 71,
GCInodeUsageThreshold: 86,
DisableTaggedMetrics: true,
BackwardsCompatibleMetrics: true,
GCInterval: 6 * time.Second,
GCParallelDestroys: 6,
GCDiskUsageThreshold: 71,
GCInodeUsageThreshold: 86,
},
Server: &ServerConfig{
Enabled: true,

2
command/agent/http.go
View File

@ -168,6 +168,8 @@ func (s *HTTPServer) registerHandlers(enableDebug bool) {
s.mux.HandleFunc("/v1/agent/servers", s.wrap(s.AgentServersRequest))
s.mux.HandleFunc("/v1/agent/keyring/", s.wrap(s.KeyringOperationRequest))
s.mux.HandleFunc("/v1/metrics", s.wrap(s.MetricsRequest))
s.mux.HandleFunc("/v1/validate/job", s.wrap(s.ValidateJobRequest))
s.mux.HandleFunc("/v1/regions", s.wrap(s.RegionListRequest))

16
command/agent/metrics_endpoint.go Normal file
View File

@ -0,0 +1,16 @@
package agent
import (
"net/http"
)
func (s *HTTPServer) MetricsRequest(resp http.ResponseWriter, req *http.Request) (interface{}, error) {
if req.Method == "GET" {
return s.newMetricsRequest(resp, req)
}
return nil, CodedError(405, ErrInvalidMethod)
}
func (s *HTTPServer) newMetricsRequest(resp http.ResponseWriter, req *http.Request) (interface{}, error) {
return s.agent.InmemSink.DisplayMetrics(resp, req)
}

52
command/agent/metrics_endpoint_test.go Normal file
View File

@ -0,0 +1,52 @@
package agent
import (
"net/http"
"net/http/httptest"
"testing"
metrics "github.com/armon/go-metrics"
"github.com/stretchr/testify/assert"
)
func TestHTTP_MetricsWithIllegalMethod(t *testing.T) {
assert := assert.New(t)
t.Parallel()
httpTest(t, nil, func(s *TestAgent) {
req, err := http.NewRequest("DELETE", "/v1/metrics", nil)
assert.Nil(err)
respW := httptest.NewRecorder()
_, err = s.Server.MetricsRequest(respW, req)
assert.NotNil(err, "HTTP DELETE should not be accepted for this endpoint")
})
}
func TestHTTP_Metrics(t *testing.T) {
assert := assert.New(t)
t.Parallel()
httpTest(t, nil, func(s *TestAgent) {
// make a separate HTTP request first, to ensure Nomad has written metrics
// and prevent a race condition
req, err := http.NewRequest("GET", "/v1/agent/self", nil)
assert.Nil(err)
respW := httptest.NewRecorder()
s.Server.AgentSelfRequest(respW, req)
// now make a metrics endpoint request, which should be already initialized
// and written to
req, err = http.NewRequest("GET", "/v1/metrics", nil)
assert.Nil(err)
respW = httptest.NewRecorder()
resp, err := s.Server.MetricsRequest(respW, req)
assert.Nil(err)
res := resp.(metrics.MetricsSummary)
gauges := res.Gauges
assert.NotEqual(0, len(gauges))
})
}

10
command/agent/testagent.go
View File

@ -13,6 +13,7 @@ import (
"strings"
"time"
metrics "github.com/armon/go-metrics"
"github.com/hashicorp/nomad/api"
"github.com/hashicorp/nomad/client/fingerprint"
"github.com/hashicorp/nomad/nomad"
@ -187,7 +188,14 @@ func (a *TestAgent) start() (*Agent, error) {
a.LogOutput = os.Stderr
}
agent, err := NewAgent(a.Config, a.LogOutput)
inm := metrics.NewInmemSink(10*time.Second, time.Minute)
metrics.NewGlobal(metrics.DefaultConfig("service-name"), inm)
if inm == nil {
return nil, fmt.Errorf("unable to set up in memory metrics needed for agent initialization")
}
agent, err := NewAgent(a.Config, a.LogOutput, inm)
if err != nil {
return nil, err
}

90
website/source/api/metrics.html.md Normal file
View File

@ -0,0 +1,90 @@
---
layout: api
page_title: Metrics - HTTP API
sidebar_current: metrics-search
description: |-
The /metrics endpoint is used to view metrics for Nomad
---
# Metrics HTTP API
The `/metrics` endpoint returns metrics for the current Nomad process.
| Method | Path | Produces |
| ------- | --------------- | -------------------------- |
| `GET` | `/v1/metrics | `application/json` |
The table below shows this endpoint's support for
[blocking queries](/api/index.html#blocking-queries) and
[required ACLs](/api/index.html#acls).
| Blocking Queries | ACL Required |
| ---------------- | ------------ |
| `NO` | `none` |
### Sample Request
```text
$ curl https://nomad.rocks/v1/metrics
```
### Sample Response
```json
{
"Counters":[
{
"Count":11,
"Labels":{},
"Max":1.0,
"Mean":1.0,
"Min":1.0,
"Name":"nomad.nomad.rpc.query",
"Stddev":0.0,
"Sum":11.0
}
],
"Gauges":[
{
"Labels":{
"node_id":"cd7c3e0c-0174-29dd-17ba-ea4609e0fd1f",
"datacenter":"dc1"
},
"Name":"nomad.client.allocations.blocked",
"Value":0.0
},
{
"Labels":{
"datacenter":"dc1",
"node_id":"cd7c3e0c-0174-29dd-17ba-ea4609e0fd1f"
},
"Name":"nomad.client.allocations.migrating",
"Value":0.0
}
],
"Samples":[
{
"Count":20,
"Labels":{},
"Max":0.03544100001454353,
"Mean":0.023678050097078084,
"Min":0.00956599973142147,
"Name":"nomad.memberlist.gossip",
"Stddev":0.005445327799243976,
"Sum":0.4735610019415617
},
{
"Count":1,
"Labels":{},
"Max":0.0964059978723526,
"Mean":0.0964059978723526,
"Min":0.0964059978723526,
"Name":"nomad.nomad.client.update_status",
"Stddev":0.0,
"Sum":0.0964059978723526
}
]
}
```

14
website/source/docs/agent/configuration/telemetry.html.md
View File

@ -58,6 +58,20 @@ The following options are available on all telemetry configurations.
- `publish_node_metrics` `(bool: false)` - Specifies if Nomad should publish
runtime metrics of nodes.
- `backwards_compatible_metrics` `(bool: false)` - Specifies if Nomad should
publish metrics that are backwards compatible with versions below 0.7, as
post version 0.7, Nomad emits tagged metrics. All new metrics will
only be added to tagged metrics. Note that this option is used to transition
monitoring to tagged metrics and will eventually be deprecated.
- `disable_tagged_metrics` `(bool: false)` - Specifies if Nomad should not emit
tagged metrics and only emit metrics compatible with versions below Nomad
0.7. Note that this option is used to transition monitoring to tagged
metrics and will eventually be deprecated.
### `statsite`
These `telemetry` parameters apply to

200
website/source/docs/agent/telemetry.html.md
View File

@ -12,7 +12,14 @@ The Nomad agent collects various runtime metrics about the performance of
different libraries and subsystems. These metrics are aggregated on a ten
second interval and are retained for one minute.
To view this data, you must send a signal to the Nomad process: on Unix,
This data can be accessed via an HTTP endpoint or via sending a signal to the
Nomad process.
Via HTTP, this data is available at `/metrics`. See
[Metrics](/api/metrics.html) for more information.
To view this data via sending a signal to the Nomad process: on Unix,
this is `USR1` while on Windows it is `BREAK`. Once Nomad receives the signal,
it will dump the current telemetry information to the agent's `stderr`.
@ -229,7 +236,196 @@ configuration block.
Please see the [agent configuration](/docs/agent/configuration/telemetry.html)
page for more details.
## Host Metrics
## Host Metrics (post Nomad version 0.7)
Starting in version 0.7, Nomad will emit tagged metrics, in the below format:
<table class="table table-bordered table-striped">
<tr>
<th>Metric</th>
<th>Description</th>
<th>Unit</th>
<th>Type</th>
<th>Labels</th>
</tr>
<tr>
<td>`nomad.client.allocated.cpu`</td>
<td>Total amount of CPU shares the scheduler has allocated to tasks</td>
<td>MHz</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.unallocated.cpu`</td>
<td>Total amount of CPU shares free for the scheduler to allocate to tasks</td>
<td>MHz</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.allocated.memory`</td>
<td>Total amount of memory the scheduler has allocated to tasks</td>
<td>Megabytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.unallocated.memory`</td>
<td>Total amount of memory free for the scheduler to allocate to tasks</td>
<td>Megabytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.allocated.disk`</td>
<td>Total amount of disk space the scheduler has allocated to tasks</td>
<td>Megabytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.unallocated.disk`</td>
<td>Total amount of disk space free for the scheduler to allocate to tasks</td>
<td>Megabytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.allocated.iops`</td>
<td>Total amount of IOPS the scheduler has allocated to tasks</td>
<td>IOPS</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.unallocated.iops`</td>
<td>Total amount of IOPS free for the scheduler to allocate to tasks</td>
<td>IOPS</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.allocated.network`</td>
<td>Total amount of bandwidth the scheduler has allocated to tasks on the
given device</td>
<td>Megabits</td>
<td>Gauge</td>
<td>node_id, datacenter, device</td>
</tr>
<tr>
<td>`nomad.client.unallocated.network`</td>
<td>Total amount of bandwidth free for the scheduler to allocate to tasks on
the given device</td>
<td>Megabits</td>
<td>Gauge</td>
<td>node_id, datacenter, device</td>
</tr>
<tr>
<td>`nomad.client.host.memory.total`</td>
<td>Total amount of physical memory on the node</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.host.memory.available`</td>
<td>Total amount of memory available to processes which includes free and
cached memory</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.host.memory.used`</td>
<td>Amount of memory used by processes</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.host.memory.free`</td>
<td>Amount of memory which is free</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.uptime`</td>
<td>Uptime of the host running the Nomad client</td>
<td>Seconds</td>
<td>Gauge</td>
<td>node_id, datacenter</td>
</tr>
<tr>
<td>`nomad.client.host.cpu.total`</td>
<td>Total CPU utilization</td>
<td>Percentage</td>
<td>Gauge</td>
<td>node_id, datacenter, cpu</td>
</tr>
<tr>
<td>`nomad.client.host.cpu.user`</td>
<td>CPU utilization in the user space</td>
<td>Percentage</td>
<td>Gauge</td>
<td>node_id, datacenter, cpu</td>
</tr>
<tr>
<td>`nomad.client.host.cpu.system`</td>
<td>CPU utilization in the system space</td>
<td>Percentage</td>
<td>Gauge</td>
<td>node_id, datacenter, cpu</td>
</tr>
<tr>
<td>`nomad.client.host.cpu.idle`</td>
<td>Idle time spent by the CPU</td>
<td>Percentage</td>
<td>Gauge</td>
<td>node_id, datacenter, cpu</td>
</tr>
<tr>
<td>`nomad.client.host.disk.size`</td>
<td>Total size of the device</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter, disk</td>
</tr>
<tr>
<td>`nomad.client.host.disk.used`</td>
<td>Amount of space which has been used</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter, disk</td>
</tr>
<tr>
<td>`nomad.client.host.disk.available`</td>
<td>Amount of space which is available</td>
<td>Bytes</td>
<td>Gauge</td>
<td>node_id, datacenter, disk</td>
</tr>
<tr>
<td>`nomad.client.host.disk.used_percent`</td>
<td>Percentage of disk space used</td>
<td>Percentage</td>
<td>Gauge</td>
<td>node_id, datacenter, disk</td>
</tr>
<tr>
<td>`nomad.client.host.disk.inodes_percent`</td>
<td>Disk space consumed by the inodes</td>
<td>Percent</td>
<td>Gauge</td>
<td>node_id, datacenter, disk</td>
</tr>
</table>
## Host Metrics (deprecated post Nomad 0.7)
The below are metrics emitted by Nomad in versions prior to 0.7. These metrics
can be emitted in the below format post-0.7 (as well as the new format,
detailed above) but any new metrics will only be available in the new format.
<table class="table table-bordered table-striped">
<tr>

4
website/source/layouts/api.erb
View File

@ -51,6 +51,10 @@
<a href="/api/nodes.html">Nodes</a>
</li>
<li<%= sidebar_current("api-metrics") %>>
<a href="/api/metrics.html">Search</a>
</li>
<li<%= sidebar_current("api-operator") %>>
<a href="/api/operator.html">Operator</a>
</li>