fix(storage): reset reclaim state after any restore, re-check owners often, share release ticks

- A restore (admin panel or RESTORE_FROM_BACKUP) now leaves
  data/.trek-reclaim-reset before it swaps the database in. Boot reconcile
  resets the reclaim state when it finds the file, whatever install id the
  database names, and removes it afterwards; a successful reopen removes it
  too. A restart after a failed reopen, or a boot restore of this
  installation's own backup, no longer keeps the old switch, marks and
  confirmations. The id mismatch reset stays.
- The release job skips a tick (install_mismatch) while the database names
  another installation or a reset is still pending.
- An allowed owner verdict is cached for 15 minutes, a refusal for 26 hours.
  The local driver reads a list prefix as a folder, so the category root is
  still listed whole, at most once per category per window.
- Each tick reads every allowed category's due rows separately, an equal
  share of the 200 each, so a category whose deletes keep failing cannot
  starve the others.
- Rows dropped after 30 days overdue and releases refused by a full queue
  are logged at warn with their counts.
This commit is contained in:
Maurice
2026-11-01 18:31:41 +01:00
parent 9d2ffc91af
commit bef8fc20f1
17 changed files with 432 additions and 80 deletions
@@ -14,9 +14,9 @@ export interface ReleaseQueueRow {
/**
* `storage_release_queue`: the objects a delete or a replace left unnamed,
* keyed by category and category-relative key, each with the time it may go.
* Owners only add rows; the release job reads the due ones of the categories
* it may judge, asks every reference source again and drops each row once it
* has decided, deleted or not. A key queued again waits from the later time.
* Owners only add rows; the release job reads the due ones of each category
* it may judge (a share of its tick each), asks every reference source again
* and drops each row once it has decided, deleted or not. A key queued again waits from the later time.
*/
export class StorageReleaseQueueRepository extends TrekRepository<StorageReleaseQueue> {
private db() {
@@ -49,15 +49,16 @@ export class StorageReleaseQueueRepository extends TrekRepository<StorageRelease
return rows.map((row) => row.category);
}
/** The rows of these categories due at `now`, the longest waiting first, at most `limit`. */
async dueIn(categories: string[], now: string, limit: number): Promise<ReleaseQueueRow[]> {
if (categories.length === 0 || limit <= 0) return [];
/** The rows of one category due at `now`, the longest waiting first, at most `limit`. */
async dueIn(category: string, now: string, limit: number): Promise<ReleaseQueueRow[]> {
if (limit <= 0) return [];
return await this.db()
.selectFrom('storage_release_queue')
.select(['category', 'key', 'due_at'])
.where('category', 'in', categories)
.where('category', '=', category)
.where('due_at', '<=', now)
.orderBy('due_at')
.orderBy('key')
.limit(limit)
.execute();
}