josh/Vetting
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

runs: add non-destructive flag + operator Cancel button
CI / Lint + build + test (push) Successful in 2m5s
Details
Release / release (push) Successful in 3m5s
Details

Browse Source 
Non-destructive pre-declares "don't touch the disks" on Start: the
Storage stage skips wipe-probe, badblocks -w, and write-mode fio,
and reports a read-only summary. Runs a new non_destructive column;
threaded through Claim → agent tests.Deps → Storage stage.

Cancel halts an in-flight run. The orchestrator transitions to a
new StateCancelled via TriggerOperatorCancelled (valid from any
active state); the agent's next heartbeat returns cmd=cancel_stage,
which fires a stored CancelFunc on the per-stage context. Stage
subprocesses spawned with exec.CommandContext die with the context,
the agent posts a cancelled outcome, then powers the host off.

Destructive stages mid-run may leave the host in an intermediate
state — the UI confirm dialog warns the operator; recovery is
manual for now.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Josh Wright
2026-04-18 13:01:42 -04:00
parent 2c440fce8a
commit 4524ab8dc0
22 changed files with 434 additions and 230 deletions
Download Patch File Download Diff File Expand all files Collapse all files
internal/store/runs.go
+19 -15
View File
@@ -14,12 +14,16 @@ type Runs struct {
DB *sql.DB
}
func (r *Runs) Create(ctx context.Context, hostID int64, tokenHash string) (int64, error) {
func (r *Runs) Create(ctx context.Context, hostID int64, tokenHash string, nonDestructive bool) (int64, error) {
now := time.Now().UTC()
nd := 0
if nonDestructive {
nd = 1
}
res, err := r.DB.ExecContext(ctx, `
INSERT INTO runs(host_id, state, agent_token_hash, next_boot_target, started_at)
VALUES(?,?,?,?,?)
`, hostID, string(model.StateQueued), tokenHash, "linux", now)
INSERT INTO runs(host_id, state, agent_token_hash, next_boot_target, started_at, non_destructive)
VALUES(?,?,?,?,?,?)
`, hostID, string(model.StateQueued), tokenHash, "linux", now, nd)
if err != nil {
return 0, fmt.Errorf("insert run: %w", err)
}
@@ -103,14 +107,14 @@ func (r *Runs) Get(ctx context.Context, id int64) (*model.Run, error) {
SELECT id, host_id, state, COALESCE(result,''), COALESCE(failed_stage,''),
COALESCE(next_boot_target,''), agent_token_hash, started_at,
completed_at, COALESCE(report_path,''), COALESCE(hold_ip,''),
COALESCE(override_flags_json,'')
COALESCE(override_flags_json,''), COALESCE(non_destructive,0)
FROM runs WHERE id = ?
`, id)
var run model.Run
var completedAt sql.NullTime
err := row.Scan(&run.ID, &run.HostID, &run.State, &run.Result, &run.FailedStage,
&run.NextBootTarget, &run.AgentTokenHash, &run.StartedAt,
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON)
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON, &run.NonDestructive)
if errors.Is(err, sql.ErrNoRows) {
return nil, ErrNotFound
}
@@ -129,7 +133,7 @@ func (r *Runs) LatestForHost(ctx context.Context, hostID int64) (*model.Run, err
SELECT id, host_id, state, COALESCE(result,''), COALESCE(failed_stage,''),
COALESCE(next_boot_target,''), agent_token_hash, started_at,
completed_at, COALESCE(report_path,''), COALESCE(hold_ip,''),
COALESCE(override_flags_json,'')
COALESCE(override_flags_json,''), COALESCE(non_destructive,0)
FROM runs WHERE host_id = ?
ORDER BY id DESC LIMIT 1
`, hostID)
@@ -137,7 +141,7 @@ func (r *Runs) LatestForHost(ctx context.Context, hostID int64) (*model.Run, err
var completedAt sql.NullTime
err := row.Scan(&run.ID, &run.HostID, &run.State, &run.Result, &run.FailedStage,
&run.NextBootTarget, &run.AgentTokenHash, &run.StartedAt,
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON)
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON, &run.NonDestructive)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
@@ -156,9 +160,9 @@ func (r *Runs) Active(ctx context.Context) ([]model.Run, error) {
SELECT id, host_id, state, COALESCE(result,''), COALESCE(failed_stage,''),
COALESCE(next_boot_target,''), agent_token_hash, started_at,
completed_at, COALESCE(report_path,''), COALESCE(hold_ip,''),
COALESCE(override_flags_json,'')
COALESCE(override_flags_json,''), COALESCE(non_destructive,0)
FROM runs
WHERE state NOT IN ('Completed','Released')
WHERE state NOT IN ('Completed','Released','Cancelled')
ORDER BY id
`)
if err != nil {
@@ -171,7 +175,7 @@ func (r *Runs) Active(ctx context.Context) ([]model.Run, error) {
var completedAt sql.NullTime
if err := rows.Scan(&run.ID, &run.HostID, &run.State, &run.Result, &run.FailedStage,
&run.NextBootTarget, &run.AgentTokenHash, &run.StartedAt,
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON); err != nil {
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON, &run.NonDestructive); err != nil {
return nil, err
}
if completedAt.Valid {
@@ -189,7 +193,7 @@ func (r *Runs) Active(ctx context.Context) ([]model.Run, error) {
func (r *Runs) CompletedOlderThan(ctx context.Context, cutoff time.Time) ([]int64, error) {
rows, err := r.DB.QueryContext(ctx, `
SELECT id FROM runs
WHERE state IN ('Completed','Released','FailedHolding')
WHERE state IN ('Completed','Released','FailedHolding','Cancelled')
AND COALESCE(completed_at, started_at) < ?
ORDER BY id
`, cutoff)
@@ -215,17 +219,17 @@ func (r *Runs) FindActiveByMAC(ctx context.Context, mac string) (*model.Run, err
SELECT r.id, r.host_id, r.state, COALESCE(r.result,''), COALESCE(r.failed_stage,''),
COALESCE(r.next_boot_target,''), r.agent_token_hash, r.started_at,
r.completed_at, COALESCE(r.report_path,''), COALESCE(r.hold_ip,''),
COALESCE(r.override_flags_json,'')
COALESCE(r.override_flags_json,''), COALESCE(r.non_destructive,0)
FROM runs r
JOIN hosts h ON h.id = r.host_id
WHERE h.mac = ? AND r.state NOT IN ('Completed','Released')
WHERE h.mac = ? AND r.state NOT IN ('Completed','Released','Cancelled')
ORDER BY r.id DESC LIMIT 1
`, mac)
var run model.Run
var completedAt sql.NullTime
err := row.Scan(&run.ID, &run.HostID, &run.State, &run.Result, &run.FailedStage,
&run.NextBootTarget, &run.AgentTokenHash, &run.StartedAt,
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON)
&completedAt, &run.ReportPath, &run.HoldIP, &run.OverrideFlagsJSON, &run.NonDestructive)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
internal/store/store_test.go
+1 -1
View File
@@ -36,7 +36,7 @@ func seedRun(t *testing.T, runs *store.Runs) (int64, int64) {
if err != nil {
t.Fatalf("create host: %v", err)
}
runID, err := runs.Create(context.Background(), hostID, "deadbeef")
runID, err := runs.Create(context.Background(), hostID, "deadbeef", false)
if err != nil {
t.Fatalf("create run: %v", err)
}