Files
coder/coderd/prebuilds/global_snapshot.go
Yevhenii Shcherbina 0f6ca55238 feat: implement scheduling mechanism for prebuilds (#18126)
Closes https://github.com/coder/internal/issues/312
Depends on https://github.com/coder/terraform-provider-coder/pull/408

This PR adds support for defining an **autoscaling block** for
prebuilds, allowing number of desired instances to scale dynamically
based on a schedule.

Example usage:
```
data "coder_workspace_preset" "us-nix" {
  ...
  
  prebuilds = {
    instances = 0                  # default to 0 instances
    
    scheduling = {
      timezone = "UTC"             # a single timezone is used for simplicity
      
      # Scale to 3 instances during the work week
      schedule {
        cron = "* 8-18 * * 1-5"    # from 8AM–6:59PM, Mon–Fri, UTC
        instances = 3              # scale to 3 instances
      }
      
      # Scale to 1 instance on Saturdays for urgent support queries
      schedule {
        cron = "* 8-14 * * 6"      # from 8AM–2:59PM, Sat, UTC
        instances = 1              # scale to 1 instance
      }
    }
  }
}
```

### Behavior
- Multiple `schedule` blocks per `prebuilds` block are supported.
- If the current time matches any defined autoscaling schedule, the
corresponding number of instances is used.
- If no schedule matches, the **default instance count**
(`prebuilds.instances`) is used as a fallback.

### Why
This feature allows prebuild instance capacity to adapt to predictable
usage patterns, such as:
- Scaling up during business hours or high-demand periods
- Reducing capacity during off-hours to save resources

### Cron specification
The cron specification is interpreted as a **continuous time range.**

For example, the expression:

```
* 9-18 * * 1-5
```

is intended to represent a continuous range from **09:00 to 18:59**,
Monday through Friday.

However, due to minor implementation imprecision, it is currently
interpreted as a range from **08:59:00 to 18:58:59**, Monday through
Friday.

This slight discrepancy arises because the evaluation is based on
whether a specific **point in time** falls within the range, using the
`github.com/coder/coder/v2/coderd/schedule/cron` library, which performs
per-minute matching rather than strict range evaluation.

---------

Co-authored-by: Danny Kopping <danny@coder.com>
2025-06-19 11:08:48 -04:00

136 lines
4.3 KiB
Go

package prebuilds
import (
"time"
"github.com/google/uuid"
"golang.org/x/xerrors"
"cdr.dev/slog"
"github.com/coder/quartz"
"github.com/coder/coder/v2/coderd/database"
"github.com/coder/coder/v2/coderd/util/slice"
)
// GlobalSnapshot represents a full point-in-time snapshot of state relating to prebuilds across all templates.
type GlobalSnapshot struct {
Presets []database.GetTemplatePresetsWithPrebuildsRow
PrebuildSchedules []database.TemplateVersionPresetPrebuildSchedule
RunningPrebuilds []database.GetRunningPrebuiltWorkspacesRow
PrebuildsInProgress []database.CountInProgressPrebuildsRow
Backoffs []database.GetPresetsBackoffRow
HardLimitedPresetsMap map[uuid.UUID]database.GetPresetsAtFailureLimitRow
clock quartz.Clock
logger slog.Logger
}
func NewGlobalSnapshot(
presets []database.GetTemplatePresetsWithPrebuildsRow,
prebuildSchedules []database.TemplateVersionPresetPrebuildSchedule,
runningPrebuilds []database.GetRunningPrebuiltWorkspacesRow,
prebuildsInProgress []database.CountInProgressPrebuildsRow,
backoffs []database.GetPresetsBackoffRow,
hardLimitedPresets []database.GetPresetsAtFailureLimitRow,
clock quartz.Clock,
logger slog.Logger,
) GlobalSnapshot {
hardLimitedPresetsMap := make(map[uuid.UUID]database.GetPresetsAtFailureLimitRow, len(hardLimitedPresets))
for _, preset := range hardLimitedPresets {
hardLimitedPresetsMap[preset.PresetID] = preset
}
return GlobalSnapshot{
Presets: presets,
PrebuildSchedules: prebuildSchedules,
RunningPrebuilds: runningPrebuilds,
PrebuildsInProgress: prebuildsInProgress,
Backoffs: backoffs,
HardLimitedPresetsMap: hardLimitedPresetsMap,
clock: clock,
logger: logger,
}
}
func (s GlobalSnapshot) FilterByPreset(presetID uuid.UUID) (*PresetSnapshot, error) {
preset, found := slice.Find(s.Presets, func(preset database.GetTemplatePresetsWithPrebuildsRow) bool {
return preset.ID == presetID
})
if !found {
return nil, xerrors.Errorf("no preset found with ID %q", presetID)
}
prebuildSchedules := slice.Filter(s.PrebuildSchedules, func(schedule database.TemplateVersionPresetPrebuildSchedule) bool {
return schedule.PresetID == presetID
})
// Only include workspaces that have successfully started
running := slice.Filter(s.RunningPrebuilds, func(prebuild database.GetRunningPrebuiltWorkspacesRow) bool {
if !prebuild.CurrentPresetID.Valid {
return false
}
return prebuild.CurrentPresetID.UUID == preset.ID
})
// Separate running workspaces into non-expired and expired based on the preset's TTL
nonExpired, expired := filterExpiredWorkspaces(preset, running)
inProgress := slice.Filter(s.PrebuildsInProgress, func(prebuild database.CountInProgressPrebuildsRow) bool {
return prebuild.PresetID.UUID == preset.ID
})
var backoffPtr *database.GetPresetsBackoffRow
backoff, found := slice.Find(s.Backoffs, func(row database.GetPresetsBackoffRow) bool {
return row.PresetID == preset.ID
})
if found {
backoffPtr = &backoff
}
_, isHardLimited := s.HardLimitedPresetsMap[preset.ID]
presetSnapshot := NewPresetSnapshot(
preset,
prebuildSchedules,
nonExpired,
expired,
inProgress,
backoffPtr,
isHardLimited,
s.clock,
s.logger,
)
return &presetSnapshot, nil
}
func (s GlobalSnapshot) IsHardLimited(presetID uuid.UUID) bool {
_, isHardLimited := s.HardLimitedPresetsMap[presetID]
return isHardLimited
}
// filterExpiredWorkspaces splits running workspaces into expired and non-expired
// based on the preset's TTL.
// If TTL is missing or zero, all workspaces are considered non-expired.
func filterExpiredWorkspaces(preset database.GetTemplatePresetsWithPrebuildsRow, runningWorkspaces []database.GetRunningPrebuiltWorkspacesRow) (nonExpired []database.GetRunningPrebuiltWorkspacesRow, expired []database.GetRunningPrebuiltWorkspacesRow) {
if !preset.Ttl.Valid {
return runningWorkspaces, expired
}
ttl := time.Duration(preset.Ttl.Int32) * time.Second
if ttl <= 0 {
return runningWorkspaces, expired
}
for _, prebuild := range runningWorkspaces {
if time.Since(prebuild.CreatedAt) > ttl {
expired = append(expired, prebuild)
} else {
nonExpired = append(nonExpired, prebuild)
}
}
return nonExpired, expired
}