Files
restic-manager/internal/store/enrollment.go
T
steve 8fb1c100fd P2-04.5: kill host.default_paths in favour of manual schedules
Two independent path lists for "what does this host back up?" was
a real divergence footgun — operator types one set at Add-host time
and a different set into a schedule, both end up in the same repo,
the snapshot history looks fine until restore. Resolution: drop
host.default_paths entirely; add a `manual` flag on schedules.
A manual schedule has paths/excludes/tags/retention like any other
but no cron — it fires only via per-schedule Run-now. Single source
of truth for what gets backed up.

Schema (migration 0007):
* schedules.manual INTEGER NOT NULL DEFAULT 0.
* For every host with non-empty default_paths, seed a manual
  schedule with those paths and bump host_schedule_version.
* ALTER TABLE hosts DROP COLUMN default_paths.
* ALTER TABLE enrollment_tokens RENAME COLUMN default_paths
  TO initial_paths.

Original draft of this migration rebuilt hosts via the
create-new + drop-old + rename-new pattern. With foreign_keys=ON
(set in the connection DSN), DROP TABLE on the parent fired
ON DELETE CASCADE on every child of hosts(id) — schedules /
jobs / snapshots / host_credentials all wiped on the smoke env
when I tried it. SQLite 3.35+ supports column-level ALTERs
directly, so we skip the rebuild dance and avoid the cascade
trap. Six lines of SQL instead of sixty, no FK risk.

Run-now rewiring:
* New `dispatchScheduleNow(hostID, scheduleID, conn?)` helper
  unifies the agent-driven path (cron fire → schedule.fire →
  OnScheduleFire callback) and the UI-driven path (operator
  clicks Run-now on a schedule row). Conn arg is optional; nil
  falls back to Hub.Send.
* New POST /hosts/{id}/schedules/{sid}/run endpoint — per-row
  Run-now button on the schedules list.
* Dashboard's per-host Run-now (handleUIRunBackup) now picks the
  host's only enabled manual schedule, falls back to the only
  enabled schedule, else returns "pick one in Schedules tab".
  Keeps one-click for the common case.

Agent:
* Scheduler skips manual schedules in cron build (silent — they're
  a normal data shape, not an error).
* Wire Schedule struct gains Manual flag.
* Schedule.fire flow unchanged — the agent only ever fires
  non-manual schedules anyway.

UI:
* Add-host form retitled "Initial schedule · manual" so the
  operator knows the paths become an editable schedule under
  the Schedules tab. Result page calls out the manual schedule
  + points at Host > Schedules.
* Schedule edit form: "Manual schedule" checkbox at the top of
  the When section; toggling it hides/shows the cron field via
  inline JS. Server-side validator skips the cron requirement
  when manual=true.
* Schedule list shows a "manual" tag under the status pill and
  renders the When column as "— run-now only —" for manual rows.
  Each row gets a Run-now button when the schedule is enabled
  and the host is online.

Tests + go test ./... green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-02 12:26:06 +01:00

132 lines
4.7 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
}
// 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
}