Complete autonomous recovery controls

This commit is contained in:
kami
2026-07-30 14:57:25 +04:00
parent 8174400b1a
commit e8fadfc998
18 changed files with 364 additions and 77 deletions
+14 -2
View File
@@ -151,6 +151,14 @@ type Task struct {
LastHarness string `json:"last_harness_id,omitempty"`
PaneState string `json:"pane_state,omitempty"` // open, closed, unreachable, unknown
LastSession SessionEvidence `json:"last_session,omitempty"`
// Recovery state is part of the durable projection, never process-local
// router memory. This makes retry and operator diagnostics survive a
// coordinator restart.
Attempt int `json:"attempt,omitempty"`
NextRetryAt time.Time `json:"next_retry_at,omitempty"`
FailureClass string `json:"failure_class,omitempty"`
LifecyclePhase string `json:"lifecycle_phase,omitempty"`
LastError string `json:"last_error,omitempty"`
}
type Event struct {
@@ -190,7 +198,7 @@ func ValidateEvent(e Event) error {
if e.SchemaVersion >= 2 && strings.TrimSpace(e.Surface) == "" {
return fmt.Errorf("%w: surface required", ErrInvalid)
}
allowed := map[string]bool{"TaskCreated": true, "TaskLeased": true, "TaskLeaseRenewed": true, "TaskReleased": true, "TaskPickupValidated": true, "TaskCompleted": true, "TaskFailed": true, "TaskBlocked": true, "TaskNeedsAttention": true, "ApprovalRequested": true, "ApprovalGranted": true, "ApprovalDenied": true, "TaskAmended": true, "TaskCorrected": true, "QuotaReported": true, "StandupAdvisory": true}
allowed := map[string]bool{"TaskCreated": true, "TaskLeased": true, "TaskLeaseRenewed": true, "TaskReleased": true, "TaskLaunchAcknowledged": true, "TaskPickupValidated": true, "TaskCompleted": true, "TaskFailed": true, "TaskBlocked": true, "TaskNeedsAttention": true, "ApprovalRequested": true, "ApprovalGranted": true, "ApprovalDenied": true, "TaskAmended": true, "TaskCorrected": true, "QuotaReported": true, "StandupAdvisory": true}
if !allowed[e.Type] {
return fmt.Errorf("%w: unknown type %q", ErrInvalid, e.Type)
}
@@ -206,7 +214,7 @@ func ValidateEvent(e Event) error {
}
if e.SchemaVersion >= 3 {
switch e.Type {
case "TaskLeased", "TaskLeaseRenewed", "TaskPickupValidated":
case "TaskLeased", "TaskLeaseRenewed", "TaskLaunchAcknowledged", "TaskPickupValidated":
if v, ok := p["lease_epoch"].(string); !ok || strings.TrimSpace(v) == "" {
return fmt.Errorf("%w: lease_epoch required", ErrInvalid)
}
@@ -313,6 +321,10 @@ func ValidatePayload(typ string, p map[string]any) error {
if err := requiredString("reason"); err != nil {
return err
}
case "TaskLaunchAcknowledged":
if err := requiredString("harness_id"); err != nil {
return err
}
case "TaskBlocked", "TaskNeedsAttention":
if err := requiredString("blocker"); err != nil {
return err
+1 -1
View File
@@ -10,7 +10,7 @@ import (
// of these on adversarial input is a defect regardless of whether real
// producers happen to send well-formed payloads.
var eventTypesUnderTest = []string{
"TaskCreated", "TaskLeased", "TaskReleased", "TaskCompleted", "TaskFailed",
"TaskCreated", "TaskLeased", "TaskReleased", "TaskLaunchAcknowledged", "TaskCompleted", "TaskFailed",
"TaskBlocked", "TaskNeedsAttention", "ApprovalRequested", "ApprovalGranted", "ApprovalDenied",
"TaskAmended", "QuotaReported", "StandupAdvisory",
}