Common render failures
The five render errors we see most, and the one-line fix for each.
### 1. Composition references a missing asset
**Signal**: render fails fast with `asset not found: /uploads/xxx.png`.
**Fix**: re-upload the asset from the chat ("upload this image again") or remove the broken `<img src>` from the composition. Assets deleted from storage stay referenced in the HTML.
### 2. Lint never went green
**Signal**: render rejected before the queue even claims it; error mentions `linter: <n> errors`.
**Fix**: ask the agent to "fix the lint errors" — it has the linter output and will patch the composition. Don't hit render again until the lint badge is green.
### 3. Render timed out (>10 min)
**Signal**: job stuck at "running" for 10+ minutes, then transitions to `failed` with `timeout`.
**Fix**: lower framerate from 60 → 30fps, or split the composition. Heavy WebGL effects (vfx-shatter, ridged-burn) over long durations are the usual culprits.
### 4. hyperframes CLI version mismatch
**Signal**: `hyperframes: unknown flag` or schema validation errors in the worker log.
**Fix**: update the worker — `./vibeedit-worker --update` or re-download from /app/settings. The cloud always runs the latest CLI; local workers can drift.
### 5. Disk full on the worker
**Signal**: `ENOSPC: no space left on device` in the worker log.
**Fix**: clear `~/.vibeedit/cache` and `~/.vibeedit/renders/tmp`. The worker doesn't auto-prune yet — long sessions accumulate intermediate frames.
### Still failing?
Email **support@vibeedit.video** with the job id from /app/renders.