Error reference (dashboard)

When something fails in the dashboard, you see a short toast explaining what happened. This page maps common toast text to its root cause and fix.

Authentication

Toast	Cause	Fix
Code is incorrect or expired	OTP wrong / older than 10 minutes	Click Resend code
Too many attempts. Request a new code	5 wrong code guesses	New code arrives in your inbox
Sign-in was cancelled or failed	OAuth flow interrupted	Retry from `/login`
We couldn’t read a verified email from that account	OAuth returned no verified email	Use a different sign-in option

Toast	Cause	Fix
Your available balance is below the 1-hour reserve	Wallet too low for the picked GPU	Top up
Settle your open invoice before launching	You have an unpaid invoice	Pay it from the billing page (card or balance)
Card declined	Stripe rejected the charge	Try a different card; check expiry
Top-up failed. Try again	Network blip during Checkout return	Retry

Toast	Cause	Fix
No machine available for this config	Nothing on the marketplace matches this spec right now	Wait a bit, or pick a different listing / GPU
That GPU count isn’t available	The machine can’t provide that many GPUs	Lower the count or pick another listing
Deploy failed	The machine couldn’t be launched	We refunded automatically; try again or pick another listing
Too many active rentals	You’re at the concurrent-rental limit	Stop one first
This rental is already in a terminal state	You clicked Stop twice	No action needed

Toast	Cause	Fix
Keys must be at least RSA 2048 or Ed25519	Pasted a weak key	`ssh-keygen -t ed25519` and retry
This public key is already on your account	Same fingerprint already saved	No action needed
Couldn’t parse that as a public key	Pasted private key by mistake, or garbled the line	Re-copy `~/.ssh/id_ed25519.pub`

Toast	Cause	Fix
Network error, try again	Browser couldn’t reach the server	Check connection; retry
Too many requests	Rate-limited	Slow down and retry in a moment

If a toast doesn’t match anything above and you’ve retried twice, screenshot it and email help@gpuoutlet.ai with:

We can correlate to the server logs via the timestamp.