Files
coder/coderd/provisionerdserver/metrics.go
T
Callum Styan 5f3be6b288 feat: add provisioner job queue wait time histogram and jobs enqueued counter (#21869)
This PR adds some metrics to help identify job enqueue rates and
latencies. This work was initiated as a way to help reduce the cost of
the observation/measurement itself for autostart scaletests, which
impacts our ability to identify/reason about the load caused by
autostart. See: https://github.com/coder/internal/issues/1209

I've extended the metrics here to account for regular user initiated
builds, prebuilds, autostarts, etc. IMO there is still the question here
of whether we want to include or need the `transition` label, which is
only present on workspace builds. Including it does lead to an increase
in cardinality, and in the case of the histogram (when not using native
histograms) that's at least a few extra series for every bucket. We
could remove the transition label there but keep it on the counter.

Additionally, the histogram is currently observing latencies for other
jobs, such as template builds/version imports, those do not have a
transition type associated with them.

Tested briefly in a workspace, can see metric values like the following:
-
`coderd_workspace_builds_enqueued_total{build_reason="autostart",provisioner_type="terraform",status="success",transition="start"}
1`
-
`coderd_provisioner_job_queue_wait_seconds_bucket{build_reason="autostart",job_type="workspace_build",provisioner_type="terraform",transition="start",le="0.025"}
1`

---------

Signed-off-by: Callum Styan <callumstyan@gmail.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-12 13:40:47 -08:00

205 lines
6.7 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
package provisionerdserver
import (
"context"
"time"
"github.com/prometheus/client_golang/prometheus"
"cdr.dev/slog/v3"
)
type Metrics struct {
logger slog.Logger
workspaceCreationTimings *prometheus.HistogramVec
workspaceClaimTimings *prometheus.HistogramVec
jobQueueWait *prometheus.HistogramVec
}
type WorkspaceTimingType int
const (
Unsupported WorkspaceTimingType = iota
WorkspaceCreation
PrebuildCreation
PrebuildClaim
)
const (
workspaceTypeRegular = "regular"
workspaceTypePrebuild = "prebuild"
)
// BuildReasonPrebuild is the build_reason metric label value for prebuild
// operations. This is distinct from database.BuildReason values since prebuilds
// use BuildReasonInitiator in the database but we want to track them separately
// in metrics. This is also used as a label value by the metrics in wsbuilder.
const BuildReasonPrebuild = workspaceTypePrebuild
type WorkspaceTimingFlags struct {
IsPrebuild bool
IsClaim bool
IsFirstBuild bool
}
func NewMetrics(logger slog.Logger) *Metrics {
log := logger.Named("provisionerd_server_metrics")
return &Metrics{
logger: log,
workspaceCreationTimings: prometheus.NewHistogramVec(prometheus.HistogramOpts{
Namespace: "coderd",
Name: "workspace_creation_duration_seconds",
Help: "Time to create a workspace by organization, template, preset, and type (regular or prebuild).",
Buckets: []float64{
1, // 1s
10,
30,
60, // 1min
60 * 5,
60 * 10,
60 * 30, // 30min
60 * 60, // 1hr
},
NativeHistogramBucketFactor: 1.1,
// Max number of native buckets kept at once to bound memory.
NativeHistogramMaxBucketNumber: 100,
// Merge/flush small buckets periodically to control churn.
NativeHistogramMinResetDuration: time.Hour,
// Treat tiny values as zero (helps with noisy near-zero latencies).
NativeHistogramZeroThreshold: 0,
NativeHistogramMaxZeroThreshold: 0,
}, []string{"organization_name", "template_name", "preset_name", "type"}),
workspaceClaimTimings: prometheus.NewHistogramVec(prometheus.HistogramOpts{
Namespace: "coderd",
Name: "prebuilt_workspace_claim_duration_seconds",
Help: "Time to claim a prebuilt workspace by organization, template, and preset.",
// Higher resolution between 15m to show typical prebuild claim times.
// Cap at 5m since longer claims diminish prebuild value.
Buckets: []float64{
1, // 1s
5,
10,
20,
30,
60, // 1m
120, // 2m
180, // 3m
240, // 4m
300, // 5m
},
NativeHistogramBucketFactor: 1.1,
// Max number of native buckets kept at once to bound memory.
NativeHistogramMaxBucketNumber: 100,
// Merge/flush small buckets periodically to control churn.
NativeHistogramMinResetDuration: time.Hour,
// Treat tiny values as zero (helps with noisy near-zero latencies).
NativeHistogramZeroThreshold: 0,
NativeHistogramMaxZeroThreshold: 0,
}, []string{"organization_name", "template_name", "preset_name"}),
jobQueueWait: prometheus.NewHistogramVec(prometheus.HistogramOpts{
Namespace: "coderd",
Name: "provisioner_job_queue_wait_seconds",
Help: "Time from job creation to acquisition by a provisioner daemon.",
Buckets: []float64{
0.1, // 100ms
0.5, // 500ms
1, // 1s
5, // 5s
10, // 10s
30, // 30s
60, // 1m
120, // 2m
300, // 5m
600, // 10m
900, // 15m
1800, // 30m
},
NativeHistogramBucketFactor: 1.1,
NativeHistogramMaxBucketNumber: 100,
NativeHistogramMinResetDuration: time.Hour,
NativeHistogramZeroThreshold: 0,
NativeHistogramMaxZeroThreshold: 0,
}, []string{"provisioner_type", "job_type", "transition", "build_reason"}),
}
}
func (m *Metrics) Register(reg prometheus.Registerer) error {
if err := reg.Register(m.workspaceCreationTimings); err != nil {
return err
}
if err := reg.Register(m.workspaceClaimTimings); err != nil {
return err
}
return reg.Register(m.jobQueueWait)
}
// IsTrackable returns true if the workspace build should be tracked in metrics.
// This includes workspace creation, prebuild creation, and prebuild claims.
func (f WorkspaceTimingFlags) IsTrackable() bool {
return f.IsPrebuild || f.IsClaim || f.IsFirstBuild
}
// getWorkspaceTimingType classifies a workspace build:
// - PrebuildCreation: creation of a prebuilt workspace
// - PrebuildClaim: claim of an existing prebuilt workspace
// - WorkspaceCreation: first build of a regular (non-prebuilt) workspace
//
// Note: order matters. Creating a prebuilt workspace is also a first build
// (IsPrebuild && IsFirstBuild). We check IsPrebuild before IsFirstBuild so
// prebuilds take precedence. This is the only case where two flags can be true.
func getWorkspaceTimingType(flags WorkspaceTimingFlags) WorkspaceTimingType {
switch {
case flags.IsPrebuild:
return PrebuildCreation
case flags.IsClaim:
return PrebuildClaim
case flags.IsFirstBuild:
return WorkspaceCreation
default:
return Unsupported
}
}
// UpdateWorkspaceTimingsMetrics updates the workspace timing metrics based on the workspace build type
func (m *Metrics) UpdateWorkspaceTimingsMetrics(
ctx context.Context,
flags WorkspaceTimingFlags,
organizationName string,
templateName string,
presetName string,
buildTime float64,
) {
m.logger.Debug(ctx, "update workspace timings metrics",
slog.F("organization_name", organizationName),
slog.F("template_name", templateName),
slog.F("preset_name", presetName),
slog.F("is_prebuild", flags.IsPrebuild),
slog.F("is_claim", flags.IsClaim),
slog.F("is_workspace_first_build", flags.IsFirstBuild))
workspaceTimingType := getWorkspaceTimingType(flags)
switch workspaceTimingType {
case WorkspaceCreation:
// Regular workspace creation (without prebuild pool)
m.workspaceCreationTimings.
WithLabelValues(organizationName, templateName, presetName, workspaceTypeRegular).Observe(buildTime)
case PrebuildCreation:
// Prebuilt workspace creation duration
m.workspaceCreationTimings.
WithLabelValues(organizationName, templateName, presetName, workspaceTypePrebuild).Observe(buildTime)
case PrebuildClaim:
// Prebuilt workspace claim duration
m.workspaceClaimTimings.
WithLabelValues(organizationName, templateName, presetName).Observe(buildTime)
default:
// Not a trackable build type (e.g. restart, stop, subsequent builds)
}
}
// ObserveJobQueueWait records the time a provisioner job spent waiting in the queue.
// For non-workspace-build jobs, transition and buildReason should be empty strings.
func (m *Metrics) ObserveJobQueueWait(provisionerType, jobType, transition, buildReason string, waitSeconds float64) {
m.jobQueueWait.WithLabelValues(provisionerType, jobType, transition, buildReason).Observe(waitSeconds)
}