Also it must be noted that an attacker having only 1.3M USD that can brute-force 72 bits (4 days of hashing on capable ASICs) can perform the same attack, so the attack is entirely feasible and no person should accept more than 1M USD using a SPV wallet.

Also the attack can be repeated: once you create the "extension point" block, you can attack more and more parties without any additional computation.

On Sat, Jun 9, 2018 at 1:03 PM Sergio Demian Lerner <sergio.d.lerner@gmail.com> wrote:

Hi Peter,

We reported this as CVE-2017-12842, although it may have been known by developers before us. 

There are hundreds of SPV wallets out there, without even considering other more sensitive systems relying on SPV proofs. 

As I said we, at RSK, discovered this problem in 2017. For RSK it's very important this is fixed because our SPV bridge uses SPV proofs.

I urge all people participating in this mailing list and the rest of the Bitcoin community to work on this issue for the security and clean-design of Bitcoin.

My suggestion is to use in the version bits 4 bits indicating the tree depth (-1), as a soft-fork, so 

00=1 depth, 

0F = 16 depth (maximum 64K transactions). Very clean.

 

The other option is to ban transaction with size lower or equal to 64. 

Best regards,

 Sergio.

On Sat, Jun 9, 2018 at 5:31 AM Bram Cohen via bitcoin-dev <bitcoin-dev@lists.linuxfoundation.org> wrote:

So are you saying that if fully validating nodes wish to prune they can maintain the ability to validate old transactions by cacheing the number of transactions in each previous block?

On Thu, Jun 7, 2018 at 3:20 PM, Peter Todd <pete@petertodd.org> wrote:

On Thu, Jun 07, 2018 at 02:15:35PM -0700, Bram Cohen wrote:
> Are you proposing a soft fork to include the number of transactions in a
> block in the block headers to compensate for the broken Merkle format? That
> sounds like a good idea.
>
> On Thu, Jun 7, 2018 at 10:13 AM, Peter Todd via bitcoin-dev <
> bitcoin-dev@lists.linuxfoundation.org> wrote:
>
> > It's well known that the Bitcoin merkle tree algorithm fails to distinguish
> > between inner nodes and 64 byte transactions, as both txs and inner nodes
> > are
> > hashed the same way. This potentially poses a problem for tx inclusion
> > proofs,
> > as a miner could (with ~60 bits of brute forcing) create a transaction that
> > committed to a transaction that was not in fact in the blockchain.
> >
> > Since odd-numbered inner/leaf nodes are concatenated with themselves and
> > hashed
> > twice, the depth of all leaves (txs) in the tree is fixed.
> >
> > It occured to me that if the depth of the merkle tree is known, this
> > vulnerability can be trivially avoided by simply comparing the length of
> > the
> > merkle path to that known depth. For pruned nodes, if the depth is saved
> > prior
> > to pruning the block contents itself, this would allow for completely safe
> > verification of tx inclusion proofs, without a soft-fork; storing this
                                         ^^^^^^^^^^^^^^^^^^^

Re-read my post: I specifically said you do not need a soft-fork to implement
this. In fact, I think you can argue that this is an accidental feature, not a
bug, as it further encourages the use of safe full verifiaction rather than
unsafe lite clients.

--
https://petertodd.org 'peter'[:-1]@petertodd.org

_______________________________________________
bitcoin-dev mailing list
bitcoin-dev@lists.linuxfoundation.org
https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev