Files
restic-manager/internal/store/enrollment.go
T
steve 02e4ef7544 testing: bootstrap UI, agent reliability, NS-01..04 + alert username
Smoothes the rough edges that came up exercising a live deployment.

First-run bootstrap UI: /bootstrap renders a username + password form
that uses the in-memory token directly (operator no longer copies it
out of the log); /login redirects there while bootstrap is available.

Agent reliability: failJob synthetic envelopes so command.run early
returns no longer hang the server-side job; runtime probe of restic
restore --help drives --no-ownership instead of version sniffing
(0.18.x had it removed). Server unit re-shaped: ProtectSystem=full
plus ReadWritePaths=/etc/restic-manager, no ProtectHome — restore
can now write anywhere a user might want.

Restore wizard: default target is /root/rm-restore/<job-id>/ with
clearer help text. Re-init confirm input uses .field (was .input,
which doesn't exist — text was invisible).

NS-01 host delete: store DeleteHost, admin-band /hosts/{id}/delete
with hostname-confirm danger zone, audit, FK cascade, live WS close.

NS-02 enrollment-token recovery: outstanding-tokens panel on
/hosts/new, regenerate (preserves attachments) and revoke handlers
+ audit, store-level ListOutstandingEnrollmentTokens and
DeleteEnrollmentToken.

NS-03 repo init / probe surface: migration 0020 adds
hosts.repo_status + repo_status_error; WS handler projects every
init job's outcome onto the host row (idempotent already-initialised
collapses to ready); creds-save resets status and dispatches a fresh
probe; /hosts/{id}/repo/probe retry endpoint with banner.

NS-04 dashboard live + sort + filter: query-string filter
(q/status/repo_status/tag/sort/dir), 5s htmx live poll mirroring the
alerts pattern with a localStorage live toggle, sortable column
headers, filter row + clear.

Alerts page: ack'd-by line resolves user_id ULID to username.

Compose.yaml ignored — host-specific.
2026-05-05 22:03:15 +01:00

247 lines
8.6 KiB
Go

package store
import (
"context"
"database/sql"
"encoding/json"
"errors"
"fmt"
"time"
)
// CreateEnrollmentToken persists a fresh one-time token. The caller
// has already hashed the raw token; the raw form is returned to the
// operator (printed in the install snippet) and never persisted.
//
// encRepoCreds is the AEAD-encrypted blob of {repo_url, repo_username,
// repo_password} that ConsumeEnrollmentToken will promote to a
// host_credentials row. Empty string = operator chose to set creds
// later via PUT /api/hosts/{id}/repo-credentials; the agent will
// refuse backup jobs until that lands.
//
// initialPaths is the JSON-encoded path list seeded into the host's
// initial manual schedule on consume. Empty string is treated as
// "[]". Not encrypted — paths aren't secret.
func (s *Store) CreateEnrollmentToken(ctx context.Context, tokenHash string, ttl time.Duration, encRepoCreds, initialPaths string) error {
now := time.Now().UTC()
var enc any = nil
if encRepoCreds != "" {
enc = encRepoCreds
}
if initialPaths == "" {
initialPaths = "[]"
}
_, err := s.db.ExecContext(ctx,
`INSERT INTO enrollment_tokens (token_hash, created_at, expires_at, enc_repo_creds, initial_paths)
VALUES (?, ?, ?, ?, ?)`,
tokenHash,
now.Format(time.RFC3339Nano),
now.Add(ttl).Format(time.RFC3339Nano),
enc, initialPaths)
if err != nil {
return fmt.Errorf("store: create enrollment token: %w", err)
}
return nil
}
// ConsumeEnrollmentToken atomically validates a token (must exist,
// not be consumed, not be expired) and marks it consumed by hostID.
// Returns ErrNotFound on any failure.
//
// The associated repo creds (if any) are promoted into
// host_credentials by the caller via SetHostCredentials *after* the
// host row exists — host_credentials has a FK to hosts that would
// otherwise fire here, since the host is created by a separate
// statement immediately after this returns.
func (s *Store) ConsumeEnrollmentToken(ctx context.Context, tokenHash, hostID string) error {
now := time.Now().UTC().Format(time.RFC3339Nano)
res, err := s.db.ExecContext(ctx,
`UPDATE enrollment_tokens
SET consumed_at = ?, consumed_host = ?
WHERE token_hash = ? AND consumed_at IS NULL AND expires_at > ?`,
now, hostID, tokenHash, now)
if err != nil {
return fmt.Errorf("store: consume enrollment token: %w", err)
}
n, _ := res.RowsAffected()
if n == 0 {
return ErrNotFound
}
return nil
}
// EnrollmentTokenAttachments is everything the enrolment handler
// needs from a token row at consume time, fetched in one round-trip.
type EnrollmentTokenAttachments struct {
// EncRepoCreds is the AEAD ciphertext bound (additional-data) to
// "token:" + token_hash. Empty if no creds were stashed.
EncRepoCreds string
// InitialPaths is the operator-supplied path list seeded into
// the host's initial manual schedule. Always non-nil (empty
// slice if none were set).
InitialPaths []string
}
// GetEnrollmentTokenAttachments returns the operator-supplied
// attachments on a still-valid enrolment token: the encrypted repo
// creds and the default-paths list. Returns ErrNotFound if the
// token is gone / consumed / expired.
//
// The caller decrypts EncRepoCreds using token_hash as AEAD
// additional data, then re-encrypts using host_id as additional
// data before passing to ConsumeEnrollmentToken.
func (s *Store) GetEnrollmentTokenAttachments(ctx context.Context, tokenHash string) (EnrollmentTokenAttachments, error) {
now := time.Now().UTC().Format(time.RFC3339Nano)
row := s.db.QueryRowContext(ctx,
`SELECT enc_repo_creds, initial_paths FROM enrollment_tokens
WHERE token_hash = ? AND consumed_at IS NULL AND expires_at > ?`,
tokenHash, now)
var (
enc sql.NullString
initialPaths string
)
if err := row.Scan(&enc, &initialPaths); err != nil {
if errors.Is(err, sql.ErrNoRows) {
return EnrollmentTokenAttachments{}, ErrNotFound
}
return EnrollmentTokenAttachments{}, fmt.Errorf("store: get enrollment token attachments: %w", err)
}
out := EnrollmentTokenAttachments{InitialPaths: []string{}}
if enc.Valid {
out.EncRepoCreds = enc.String
}
if initialPaths != "" {
_ = json.Unmarshal([]byte(initialPaths), &out.InitialPaths)
}
return out, nil
}
// EnrollmentTokenStatus is what the awaiting-agent panel polls for
// after Add-host. Returned by GetEnrollmentTokenStatus; the
// consuming code branches on Consumed + the (optional) ConsumedHost.
type EnrollmentTokenStatus struct {
ExpiresAt time.Time
ConsumedAt *time.Time
ConsumedHost *string
}
// GetEnrollmentTokenStatus reports whether a token has been
// consumed yet (the agent has called /api/agents/enroll). Returns
// ErrNotFound if the token is unknown — the polling endpoint maps
// that to "token expired or invalid; stop polling".
func (s *Store) GetEnrollmentTokenStatus(ctx context.Context, tokenHash string) (EnrollmentTokenStatus, error) {
row := s.db.QueryRowContext(ctx,
`SELECT expires_at, consumed_at, consumed_host
FROM enrollment_tokens WHERE token_hash = ?`,
tokenHash)
var (
expiresAt string
consumedAt, host sql.NullString
)
if err := row.Scan(&expiresAt, &consumedAt, &host); err != nil {
if errors.Is(err, sql.ErrNoRows) {
return EnrollmentTokenStatus{}, ErrNotFound
}
return EnrollmentTokenStatus{}, fmt.Errorf("store: get enrollment token status: %w", err)
}
out := EnrollmentTokenStatus{}
if t, err := time.Parse(time.RFC3339Nano, expiresAt); err == nil {
out.ExpiresAt = t
}
if consumedAt.Valid {
if t, err := time.Parse(time.RFC3339Nano, consumedAt.String); err == nil {
out.ConsumedAt = &t
}
}
if host.Valid {
s := host.String
out.ConsumedHost = &s
}
return out, nil
}
// OutstandingEnrollmentToken is what the recoverable-token list page
// shows: enough to identify the row (short hash + created/expires)
// and re-render the install snippet via the regenerate flow, plus
// the encrypted repo creds blob the caller can decrypt-and-redact for
// display.
type OutstandingEnrollmentToken struct {
TokenHash string
CreatedAt time.Time
ExpiresAt time.Time
EncRepoCreds string
InitialPaths []string
}
// ListOutstandingEnrollmentTokens returns every still-valid token
// (un-consumed and not expired). Used by the Add-host page to give
// operators a way back to the install snippet after they close the
// /hosts/pending/{token} tab without finishing onboarding.
func (s *Store) ListOutstandingEnrollmentTokens(ctx context.Context) ([]OutstandingEnrollmentToken, error) {
now := time.Now().UTC().Format(time.RFC3339Nano)
rows, err := s.db.QueryContext(ctx,
`SELECT token_hash, created_at, expires_at, enc_repo_creds, initial_paths
FROM enrollment_tokens
WHERE consumed_at IS NULL AND expires_at > ?
ORDER BY created_at DESC`, now)
if err != nil {
return nil, fmt.Errorf("store: list outstanding enrollment tokens: %w", err)
}
defer func() { _ = rows.Close() }()
var out []OutstandingEnrollmentToken
for rows.Next() {
var (
hash, created, expires string
enc sql.NullString
pathsJSON string
)
if err := rows.Scan(&hash, &created, &expires, &enc, &pathsJSON); err != nil {
return nil, fmt.Errorf("store: scan outstanding enrollment token: %w", err)
}
row := OutstandingEnrollmentToken{TokenHash: hash, InitialPaths: []string{}}
if t, err := time.Parse(time.RFC3339Nano, created); err == nil {
row.CreatedAt = t
}
if t, err := time.Parse(time.RFC3339Nano, expires); err == nil {
row.ExpiresAt = t
}
if enc.Valid {
row.EncRepoCreds = enc.String
}
if pathsJSON != "" {
_ = json.Unmarshal([]byte(pathsJSON), &row.InitialPaths)
}
out = append(out, row)
}
return out, rows.Err()
}
// DeleteEnrollmentToken removes a token row. Used by the operator-
// driven revoke flow and by regenerate (which deletes the old hash
// then mints a fresh one). Idempotent: ErrNotFound on miss.
func (s *Store) DeleteEnrollmentToken(ctx context.Context, tokenHash string) error {
res, err := s.db.ExecContext(ctx,
`DELETE FROM enrollment_tokens WHERE token_hash = ?`, tokenHash)
if err != nil {
return fmt.Errorf("store: delete enrollment token: %w", err)
}
n, _ := res.RowsAffected()
if n == 0 {
return ErrNotFound
}
return nil
}
// PurgeExpiredEnrollmentTokens deletes long-expired token rows. Tokens
// retained for ~24h after expiry so audit traces still resolve them.
func (s *Store) PurgeExpiredEnrollmentTokens(ctx context.Context) (int64, error) {
cutoff := time.Now().Add(-24 * time.Hour).UTC().Format(time.RFC3339Nano)
res, err := s.db.ExecContext(ctx,
`DELETE FROM enrollment_tokens WHERE expires_at <= ?`, cutoff)
if err != nil {
return 0, fmt.Errorf("store: purge enrollment tokens: %w", err)
}
n, _ := res.RowsAffected()
return n, nil
}