Skip to main content

Troubleshooting

Every failure gives you a stable code and a request-id that ties straight to the server log, on-chain spends are race-guarded so you never double-pay, and clear pre-flight guards stop wasted capsules before you spend $DIG.

The mental model — find your failure by its code​

Every surface — the dig RPC, the digstore CLI, DIGHUb, the chia:// loader, the SDK — maps a failure to one STABLE code. Branch on the code, never the message. One consolidated catalog covers all of them and is also published machine-readable.

Pre-flight guards (digstore doctor, --dry-run, --if-changed) and resumable anchors mean a stuck or no-op publish never silently spends.

Common publishing failures​

Insufficient funds, a confirm timeout (resumable — your spend isn't lost), and the non-fast-forward "remote root has advanced".

→ Troubleshooting

Read & verify failures​

Proof mismatch, decrypt/salt errors, and not-found / decoy responses.

→ Read & verify failures

Wallet & session issues​

Connect, re-auth, a declined request, and watch-only sessions that can't sign.

→ Wallet session can't sign

Pre-flight & cost checks — don't waste a capsule​

digstore doctor (environment + readiness), --dry-run (preview the cost and the would-be capsule), and --if-changed (a byte-identical build is a no-op).

→ Deploy from GitHub Actions · On-chain anchoring → cost & safety

Error codes reference​

CLI exit codes · RPC -32xxx · DIGHUb · dig-loader · SDK — one consolidated table.

→ Error codes

FAQ​

Cost, the free trial, why the price is uniform, where to get $DIG, and "is there a testnet?".

→ FAQ

Get help​

Discord + GitHub, and how to file a good report — never paste secrets.

→ Get help

Status & changelog​

→ Status · Changelog

Go deeper: the protocol​