Skip to content
Running your server

Automation

Automation ​

Set it up once and BackIssue keeps your collection current on its own.

Scheduled jobs ​

System → Jobs shows every schedule with its cron expression, an enable toggle, last-run result, and a Run now button.

The top of System → Jobs. Each row is one schedule: toggle, cron pattern, when it runs next, and Run now. 1 A job runs one instance at a time, so a running one can't be started again. 2 Shipped-off lanes already carry a cron — flick the toggle and they start.
JobWhat it doesShips
Check this week's releasesFetches this week's releases and flags new issues of your monitored seriesOn, twice daily
Back up databaseSnapshots catalog.db into backups/ (keeps the newest 5) — cheap insurance for the database that holds your collection, accounts, and reading historyOn, weekly
Match series metadata (owned + followed)Matches any unmatched series to a metadata record (useful after imports)Off
Watch indexer RSS for releasesPolls your indexers' latest uploads feed (usenet + torrents) and grabs anything that matches a wanted issue — a new upload is caught within one poll instead of waiting for the next search. Each upload is considered exactly onceOff
Search new releasesThe fast lane for this week's comics: queues wanted issues released in the last 14 days (the recentSearchDays key in settings.json, 1–90; it has no field in Settings). Unlike the backfill, failures are retried on every run while the issue is inside the window — new releases reach indexers over days — then age outOff
Search wanted issues (backfill)The patient backfill: works through the whole Wanted list in batches, skipping anything in flight or previously failed, so back-catalog gaps fill steadily without hammering sourcesOff
Grab weekly 0-Day pack (torrent)Torrents only: grabs the weekly 0-day pack and imports just your gaps — see Download sourcesOff

Everything except the releases check and the database backup ships switched off — turn on the lanes you want. Each already carries a sensible cron, so usually all you do is flick the toggle.

Plugins add jobs of their own, and they appear in the same list once the plugin is installed:

JobFrom
Watch AirDC++ announcementsAirDC++
Scan book libraries, Fill wanted booksBooks
Scan audiobook libraries, Sync remote audiobook catalog, Fill wanted audiobooksAudiobooks
Fill approved book requestsRequests
Rebuild the browse indexShelves
Pre-render issue covers (On, hourly)Reader

The three search lanes are complementary layers: RSS watch reacts to uploads in minutes, new releases actively hunts the current window (and retries), and the wanted backfill chews the backlog nightly. Anything one misses, the next catches.

Schedules use cron expressions (0 3 * * * = daily at 03:00). Each job runs at most one instance at a time, and every run is recorded with its outcome. A run missed while the app was off catches up once at the next start — not once per missed slot.

Suggested starting points:

Check this week's releases                0 8 * * 3       # Wednesday mornings (new comic day)
Match series metadata (owned + followed)  0 4 * * *       # nightly
Watch indexer RSS for releases            */15 * * * *    # every 15 minutes (off until you enable it)
Search new releases                       0 */6 * * *     # every 6 hours (off until you enable it)
Search wanted issues (backfill)           0 2 * * *       # nightly, batch of 25–50
Grab weekly 0-Day pack (torrent)          0 9 * * 3,4     # Wed/Thu — packs appear midweek

Notifications ​

BackIssue has a built-in notification centre — the 🔔 bell in the header. It records notable events and shows an unread badge; open it for recent notifications, mark them read, or click one to jump to what it's about. It updates live, so things appear as they happen.

Notifications are per user — some are broadcast to everyone (a download finished, a release-day heads-up), others are targeted (your request was approved). What you see:

  • Downloads completed or failed, packs imported or failed
  • A weekly release-day heads-up for series you follow
  • Requests activity — a new request filed (for reviewers), or your request approved/declined

Sending events elsewhere ​

The Notifications Hub plugin sends the same events to Discord (rich embeds with cover art), Telegram, Pushover, ntfy, or any webhook — each channel with its own category filter. The in-app bell always records everything regardless.

History ​

Sidebar → History — every import, forever: when, which series/issue, which source served it, and where the file went. The paper trail for "where did this file come from?"

It has three views:

  • Imported — what landed, newest first.
  • Failed — grabs that didn't make it, each with the reason the source or the client gave.
  • Blocklist — releases that failed badly enough to be barred from automatic grabs, with why. Remove one to let it be grabbed again, or clear the lot. A manual source search is never filtered by it.
History in its default view: imports grouped by day, each tagged with the source that served it — hover a row in the app for where the file went. The Failed and Blocklist chips switch to the other two views.

Logs ​

System → Logs — timestamped, filterable application log: searches, grabs, imports, tag results, failures with reasons. First stop when something didn't download; the reason is almost always spelled out here.

Stats ​

Sidebar → Stats is the read-only health check for the whole collection — size on disk, file and format counts, completion, metadata health and download activity. It is described in full under Stats.

Live updates ​

Everything above updates in real time — the UI holds an event stream to the server, so queue progress, badge counts, job status, and release ownership all move without refreshing. If the stream drops (sleeping laptop, network blip), the UI falls back to polling and recovers on its own.