fix: workers stuck, images not processing after restart, dashboard stats frozen#22
Open
vonhex wants to merge 4 commits intoSpaceinvaderOne:mainfrom
Open
fix: workers stuck, images not processing after restart, dashboard stats frozen#22vonhex wants to merge 4 commits intoSpaceinvaderOne:mainfrom
vonhex wants to merge 4 commits intoSpaceinvaderOne:mainfrom
Conversation
- watcher: fix _scan() returning 2-tuple instead of 3-tuple when photos dir is missing — caused ValueError crash on every initial scan if the mount wasn't ready, silently killing image discovery - main: on startup, reset processing→pending and re-enqueue all pending images so images stranded from a previous session aren't stuck forever - worker: add resize() to gracefully stop/restart workers with a new count - settings: call worker.resize() when concurrent_workers changes so the live pool count actually updates Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ulating, stop/resume flow
- app.js: move `const w = data.worker || {}` before its first use — it was
referenced at line 175 but defined at line 227, causing a ReferenceError
that silently killed the entire updateDashboard callback (worker status
stayed 'Loading...', progress stayed 'Calculating...')
- app.js: replace the disabled 'Stopping...' button state with an enabled
'Resume Processing' button so users can restart after stopping without
needing a page reload or app restart
- scan.py: add POST /api/scan/resume — clears stop_requested and re-enqueues
all pending images so processing can continue after a stop
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a one-click way to re-queue every failed image without having to select them manually page by page. - POST /api/images/retry-all-errors — resets all error images to pending (clears error_message) and enqueues them all in one DB query - Queue template: Retry All Errors button shown alongside Retry Selected when the error filter is active - app.js: retryAllErrors() function that calls the endpoint and reloads the page with a count toast Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Author
|
also made a retry app button for errors so you dont have to go page by page to requeue |
…adaptive polling - database: enable SQLite WAL mode + NORMAL sync + 32MB cache — readers and writers no longer block each other; eliminates the main source of UI stalls when the worker is actively processing images - thumbnails: move PIL thumbnail generation into asyncio.to_thread() so CPU/IO-bound image work no longer blocks the event loop and stalls every other concurrent request - images: thumbnail endpoint now fetches only file_path (not SELECT *) and returns Cache-Control: max-age=86400, immutable so the browser caches thumbnails for 24h instead of re-fetching on every page visit - dashboard: back off polling to 15s when idle, stay at 5s when active — halves the steady-state DB query rate when nothing is processing Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Three bugs that cause images to never process and the dashboard to show stale data:
Scan crash on missing/slow mount — `_scan()` returned a 2-tuple `(0, 0)` but all callers unpack 3 values `(new_count, skipped, new_ids)`. If `/photos` isn't ready at startup (slow network mount), this threw a silent `ValueError` that killed the initial scan entirely.
Pending images stuck after restart — On restart the worker queue starts empty. The scanner skips files already in the DB (`skip_processed=True`), so any images that were `pending` or mid-flight (`processing`) from a previous session are stranded forever. Fixed by resetting `processing → pending` on startup and re-enqueuing all `pending` rows before the watcher starts.
Dashboard frozen at "Loading..." / "Calculating..." — In `updateDashboard()`, `w.stop_requested` was referenced before `const w = data.worker || {}` was declared, throwing a `ReferenceError` silently swallowed by `.catch(() => {})`. The worker bar never left "Loading..." and the progress text never left "Calculating...". Also adds a Resume Processing button wired to a new `POST /api/scan/resume` endpoint so users can restart after stopping without an app restart.
Test plan
🤖 Generated with Claude Code