Repository navigation
[Bug]: Failed to fetch dynamically imported module #10613
Description
Before submitting
- I searched existing issues and did not find a duplicate.
- I included enough detail to reproduce or investigate the problem.
Area
apps/desktop
Steps to reproduce
- used app in nightly 0.0.40 (idk the exact version number)
- i saw the update button and wanted to update. this did not work due to some electron issue (didnt pay close attention). i also noticed that the openai provider was not working.
- decided to switch to stable
- after relauch i saw the error from section "Actual Behavior"
Expected behavior
t3 code (alpha) launches normally
Actual behavior
t3 code (alpha) launches but throws the following error:
T3 Code (Alpha) 0.0.39
Path: /settings/general
Time: 2026-09-07T23:13:05.355Z
TypeError: Failed to fetch dynamically imported module: t3code://app/assets/settings-C9snXPCM.js
Impact
Minor bug or occasional failure
Version or commit
No response
Environment
0.0.39 Desktop, Stable
Logs or stack traces
Screenshots, recordings, or supporting files
No response
Workaround
I retried a couple minutes later and it launched just fine.
Activity
JDeffner commentedon Sep 7, 2026
after working just fine i went to settings which worked just fine but after clicking on providers i got another error:
T3 Code (Alpha) 0.0.39
Path: /settings/providers
Time: 2026-09-07T23:22:28.185Z
TypeError: Failed to fetch dynamically imported module: t3code://app/assets/settings.providers-OJREr2We.js
after waiting some time and clicking reload i was able to open the settings again.
juliusmarminge commentedon Sep 7, 2026
Thanks for the report — this is enough to triage.
What happened
You switched Nightly (~0.0.40) → Stable 0.0.39 and the first launch hit the settings route. That page is a Vite-hashed split chunk (
t3code://app/assets/settings-….js). After a channel switch, that fetch can miss once: Chromium/module graph still pointing at the previous build’s hash, or the local static server not serving the new assets yet. The overlay you pasted is the root route error view. A later retry loading cleanly matches a one-shot stale-chunk / first-boot race, not a broken Stable install.Not the same as Nightly→Stable project/chat hiding
#4374 and closed #3840 are about projects/threads disappearing after a downgrade. This issue is a transient dynamic-import failure on
/settings/generalthat recovered on retry.What we already have
#9058 added a one-shot reload on
vite:preloadError(chunkReloadGuard). Route-chunk failures can still land in the error boundary instead of that reload path, which is why you saw the overlay.Workaround
Reload / relaunch (what you did). If it comes back after a reload, copy a fresh error report.
Next
Keep this open as a minor desktop/web bug: handle “Failed to fetch dynamically imported module” on the settings (and other cold) route chunks with the same guarded reload, so a Nightly→Stable relaunch does not show this overlay.
[-][Bug]: Error after changing from Nightly to Stable[/-][+][Bug]: Failed to fetch dynamically imported module[/+]juliusmarminge commentedon Nov 12, 2026
@JDeffner Which OS and install type were you on (Windows installer, macOS app, Linux AppImage or .deb), and did you switch to Stable through the update track in Settings (letting the app download and install it) or by installing the Stable download yourself?
JDeffner commentedon Nov 13, 2026
This was the installed Windows desktop app (.exe installer). My machine reports Windows 10 Home, 64-bit; I cannot recover the exact OS build from the incident.
I switched from Nightly to Stable through the update track in Settings, rather than downloading Stable manually. After relaunch, Stable 0.0.39 showed the dynamic-import errors reported above. Waiting and reloading restored access. The original updater logs are no longer retained, so I cannot recover the exact Nightly build or the initial Electron update error.