You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, the wallet's seed is being generated directly from BIP39 entropy, whereas it should be being fed through 2048 rounds of PBKDF2 with a password (even if blank). This should be rectified for new wallets (with a simple call to 'to seed'), with a mechanism to ensure older mnemonics can still be used to restore.
The text was updated successfully, but these errors were encountered:
As discussed in the wallet dev channel, this is unfortunate that we're not properly adhering to the BIP39 spec, but there is no way to migrate wallets to this change without a severe mess. In short:
Both the entropy and PBKDF2(entropy) are equally valid wallet seeds. If a user were to attempt to restore from a phrase in a wallet that had support for PBKDF2 + Legacy, a restore would need to scan the entire chain for both versions to determine which that particular seed should support. This would become even messier if a user recovered from seed with a phrase, then did a few transactions, then did a further check to find older output.
There are no particular security benefits to the PBKDF2 version over the entropy (we're not using a BIP39 passphrase)
We do need to document the recovery process somewhere, particularly as it's non standard. I'll open a separate issue for that.
Currently, the wallet's seed is being generated directly from BIP39 entropy, whereas it should be being fed through 2048 rounds of PBKDF2 with a password (even if blank). This should be rectified for new wallets (with a simple call to 'to seed'), with a mechanism to ensure older mnemonics can still be used to restore.
The text was updated successfully, but these errors were encountered: