molecule-ai/molecule-core
35
0
Fork 2
Code Issues 16 Pull Requests 2 Actions 49 Packages Projects Releases Wiki Activity

Merge branch 'staging' into fix/coverage-gate-platform-go-1823

Browse Source
This commit is contained in:
molecule-ai[bot] 2026-04-23 18:40:23 +00:00 committed by GitHub
commit bbc59fccf8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 60 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
workspace-server/internal/handlers/registry.go
View File

@ -451,16 +451,17 @@ func (h *RegistryHandler) evaluateStatus(c *gin.Context, payload models.Heartbea
h.broadcaster.RecordAndBroadcast(ctx, "WORKSPACE_ONLINE", payload.WorkspaceID, map[string]interface{}{})
}
// Auto-recovery: if a workspace is marked "failed" or "provisioning" but is
// actively sending heartbeats, it has clearly booted successfully. Transition
// to "online" so the scheduler and dashboard reflect reality. This catches
// cases where the provisioner crashed mid-setup or an earlier error left the
// status stale.
if currentStatus == "failed" || currentStatus == "provisioning" {
if _, err := db.DB.ExecContext(ctx, `UPDATE workspaces SET status = 'online', updated_at = now() WHERE id = $1 AND status IN ('failed', 'provisioning')`, payload.WorkspaceID); err != nil {
log.Printf("Heartbeat: failed to auto-recover %s from %s to online: %v", payload.WorkspaceID, currentStatus, err)
// Auto-recovery: if a workspace is marked "provisioning" but is actively sending
// heartbeats, it has successfully started up. Transition to "online" so the scheduler
// and A2A proxy can dispatch tasks to it. The provisioner does not call
// /registry/register on container start — only the heartbeat loop does, so this
// transition is the only mechanism that moves newly-started workspaces out of
// the phantom-idle state. (#1784)
if currentStatus == "provisioning" {
if _, err := db.DB.ExecContext(ctx, `UPDATE workspaces SET status = 'online', updated_at = now() WHERE id = $1 AND status = 'provisioning'`, payload.WorkspaceID); err != nil {
log.Printf("Heartbeat: failed to transition %s from provisioning to online: %v", payload.WorkspaceID, err)
} else {
log.Printf("Heartbeat: auto-recovered %s from %s to online (heartbeat received)", payload.WorkspaceID, currentStatus)
log.Printf("Heartbeat: transitioned %s from provisioning to online (heartbeat received)", payload.WorkspaceID)
}
h.broadcaster.RecordAndBroadcast(ctx, "WORKSPACE_ONLINE", payload.WorkspaceID, map[string]interface{}{
"recovered_from": currentStatus,

50
workspace-server/internal/handlers/registry_test.go
View File

@ -134,6 +134,56 @@ func TestHeartbeatHandler_OfflineToOnline(t *testing.T) {
}
}
// ==================== Heartbeat — provisioning → online recovery (#1784) ====================
func TestHeartbeatHandler_ProvisioningToOnline(t *testing.T) {
mock := setupTestDB(t)
setupTestRedis(t)
broadcaster := newTestBroadcaster()
handler := NewRegistryHandler(broadcaster)
// Expect prevTask SELECT
mock.ExpectQuery("SELECT COALESCE\\(current_task").
WithArgs("ws-provisioning").
WillReturnRows(sqlmock.NewRows([]string{"current_task"}).AddRow(""))
// Expect heartbeat UPDATE
mock.ExpectExec("UPDATE workspaces SET").
WithArgs("ws-provisioning", 0.0, "", 1, 3000, "").
WillReturnResult(sqlmock.NewResult(0, 1))
// Expect evaluateStatus SELECT — currently provisioning
mock.ExpectQuery("SELECT status FROM workspaces WHERE id =").
WithArgs("ws-provisioning").
WillReturnRows(sqlmock.NewRows([]string{"status"}).AddRow("provisioning"))
// Expect status transition to online (#1784)
mock.ExpectExec("UPDATE workspaces SET status = 'online'").
WithArgs("ws-provisioning").
WillReturnResult(sqlmock.NewResult(0, 1))
// Expect RecordAndBroadcast INSERT for WORKSPACE_ONLINE
mock.ExpectExec("INSERT INTO structure_events").
WillReturnResult(sqlmock.NewResult(0, 1))
w := httptest.NewRecorder()
c, _ := gin.CreateTestContext(w)
body := `{"workspace_id":"ws-provisioning","error_rate":0.0,"sample_error":"","active_tasks":1,"uptime_seconds":3000}`
c.Request = httptest.NewRequest("POST", "/registry/heartbeat", bytes.NewBufferString(body))
c.Request.Header.Set("Content-Type", "application/json")
handler.Heartbeat(c)
if w.Code != http.StatusOK {
t.Errorf("expected status 200, got %d: %s", w.Code, w.Body.String())
}
if err := mock.ExpectationsWereMet(); err != nil {
t.Errorf("unmet sqlmock expectations: %v", err)
}
}
func TestHeartbeatHandler_BadJSON(t *testing.T) {
setupTestDB(t)
setupTestRedis(t)