Instead of forwarding every error, we deliberately ignore certain
variants that are not worth being printed to the log. In particular,
this concerns "UnsupportedProtocols" and "ResponseOmission".
To make this less verbose we introduce a macro for mapping a
`RequestResponseEvent` to `{alice,bob}::OutEvent`. We use a macro
because those `OutEvent`s are different types and the only other
way of abstracting over them would be to introduce traits that we
implement on both of them.
To make the macro easier to use, we move all the `From` implementations
that convert between the protocol and the more high-level behaviour
into the actual protocol module.
405: Concurrent swaps with same peer r=da-kami a=da-kami
Fixes#367
- [x] Concurrent swaps with same peer
Not sure how much more time I should invest into this. We could just merge the current state and then do improvements on top...?
Improvements:
- [x] Think `// TODO: Remove unnecessary swap-id check` through and remove it
- [x] Add concurrent swap test, multiple swaps with same Bob
- [ ] Save swap messages without matching swap in execution in the database
- [ ] Assert the balances in the new concurrent swap tests
- [ ] ~~Add concurrent swap test, multiple swaps with different Bobs~~
- [ ] ~~Send swap-id in separate message, not on top of `Message0`~~
Co-authored-by: Daniel Karzel <daniel@comit.network>
- Swap-id is exchanged during execution setup. CLI (Bob) sends the swap-id to be used in his first message.
- Transfer poof and encryption signature messages include the swap-id so it can be properly associated with the correct swap.
- ASB: Encryption signatures are associated with swaps by swap-id, not peer-id.
- ASB: Transfer proofs are still associated to peer-ids (because they have to be sent to the respective peer), but the ASB can buffer multiple
- CLI: Incoming transfer proofs are checked for matching swap-id. If a transfer proof with a different swap-id than the current executing swap is received it will be ignored. We can change this to saving into the database.
Includes concurrent swap tests with the same Bob.
- One test that pauses and starts an additional swap after the transfer proof was received. Results in both swaps being redeemed after resuming the first swap.
- One test that pauses and starts an additional swap before the transfer proof is sent (just after BTC locked). Results in the second swap redeeming and the first swap being refunded (because the transfer proof on Bob's side is lost). Once we store transfer proofs that we receive during executing a different swap into the database both swaps should redeem.
Note that the monero harness was adapted to allow creating wallets with multiple outputs, which is needed for Alice.
It might very well be that the cancel transaction is already published.
If that is the case, there is no point in failing the command. We simply
transition to cancel and exit normally.
The reason this comes up now is because Alice now properly waits for
the cancel timelock as well and publishes the cancel transaction first.
Ultimately, she should not do that because there is no benefit to her
unless she can also publish the punish transaction.
We use the "precondition" feature of the `tokio::select!` macro to
avoid polling certain futures. In particular, we skip polling all
futures that - when resolved - require us to send a message to Alice.
bmrng is a library providing a request-response channel that allows
the receiving end of the channel to send a response back to the sender.
This allows us to more accurately implement the functions on the
`EventLoopHandle`. In particular, we now _wait_ for the ACK of specific
messages from the other party before resolving the future.
For example, when sending the encrypted signature, the async function
on the `EventLoopHandle` does not resolve until we received the ACK
from the other party.
We also delete the `Channels` abstraction in favor of directly creating
bmrng channels. This allows us to directly control the channel buffer
which we set to 1 because we don't need more than that on Bob's side.
The execution setup is our only libp2p protocol that doesn't have
a timeout built-in. Hence, if anything fails on Alice's side, we
would wait here forever.
Wrapping the future in a timeout ensures that we fail eventually
if this protocol doesn't succeed.
We don't need to hide the fields of this Behaviour as the only reason
for why this struct exists is because libp2p forces us to compose our
NetworkBehaviours into a new struct.
In order for the re-construction of TxLock to be meaningful, we limit
`Message2` to the PSBT instead of the full struct. This is a breaking
change in the network layer.
The PSBT is valid if:
- It has at most two outputs (we allow a change output)
- One of the outputs pays the agreed upon amount to a shared output script
Resolves#260.
This allows us to remove all visibility modifiers from the message
fields because child modules (in this case {alice,bob}::state) can
always access private fields of structs.
It also moves the messages into a more natural place. Previously,
they were defined within the network layer even though they are
independent of the libp2p implementation.
Instead of watching for status changes directly on bitcoin::Wallet,
we return a Subscription object back to the caller. This subscription
object can be re-used multiple times.
Among other things, this now allows callers of `broadcast` to decide
on what to wait for given the returned Subscription object.
The new API is also more concise which allows us to remove some of
the functions on the actor states in favor of simple inline calls.
Co-authored-by: rishflab <rishflab@hotmail.com>
The swap should not be concerned with connection handling. This is
the responsibility of the overall application.
All but the execution-setup NetworkBehaviour are `request-response`
behaviours. These have built-in functionality to automatically emit
a dial attempt in case we are not connected at the time we want to
send a message. We remove all of the manual dialling code from the
swap in favor of this behaviour.
Additionally, we make sure to establish a connection as soon as the
EventLoop gets started. In case we ever loose the connection to Alice,
we try to re-establish it.
To achieve this, we decompose `watch_for_locked_xmr` into two parts:
1. A non-self-consuming function to construct a `WatchRequest`
2. A state transition that can now consume `self` again because
it is only called once within the whole select! expression.
Ideally, we would move more logic onto this state transition (like
comparing the actual amounts and fail the transition if it is not
valid). Doing so would have an unfortunate side-effect: We would
always wait for the full confirmations before checking whether or
not we actually receive enough XMR.
This allows us to have state transitions that consume self.
Instead of calling this function in all the branches, we can simply
make the whole match statement evaluate to the new state and perform
this functionality at the very end.
322: Refactor `ExecutionParams` and harmonize sync intervals of wallets r=thomaseizinger a=thomaseizinger
Co-authored-by: Thomas Eizinger <thomas@eizinger.io>
Bob does not care whether tx lock is confirmed. That is alice's problem.
This wait was introduced to remedy a bug in status_of_script() which was
failing when called on a transaction with no confirmations.
We have a repeated pattern where we construct one of our
Tx{Cancel,Redeem,Punish,Refund,Lock} transactions and wait until
the status of this transaction changes. We can make this more
ergonomic by creating and implementing a `Watchable` trait that
gives access to the TxId and relevant script for this transaction.
This allows us to remove a parameter from the `watch_until_status`
function.
Additionally, there is a 2nd pattern: "Completing" one of these
transaction and waiting until they are confirmed with the configured
number of blocks for finality. We can make this more ergonomic by
returning a future from `broadcast` that callers can await in case
they want to wait for the broadcasted transaction to reach finality.
The execution params don't change throughout the lifetime of the
program. They can be set in the wallet at the very beginning.
This simplifies the interface of the wallet functions.
We achieve our optimizations in three ways:
1. Batching calls instead of making them individually.
To get access to the batch calls, we replace all our
calls to the HTTP interface with RPC calls.
2. Never directly make network calls based on function
calls on the wallet.
Instead, inquiring about the status of a script always
just returns information based on local data. With every
call, we check when we last refreshed the local data and
do so if the data is considered to be too old. This
interval is configurable.
3. Use electrum's notification feature to get updated
with the latest blockheight.
Co-authored-by: Thomas Eizinger <thomas@eizinger.io>
Co-authored-by: Rishab Sharma <rishflab@hotmail.com>
First, we tell the user that we are now waiting for Alice to lock
the monero. Additionally, we tell them once we received the
transfer proof which will lead directly into the
"waiting for confirmations" function.
271: Bob can verify that the XMR lock tx was published r=da-kami a=da-kami
The Monero `txhash` log was removed. I feel the user should have the possibility to verify that the transaction was actually published so I added the tx-hash to the confirmation output.
We could potentially print the tx-hash when receiving the transfer proof already, but that might not add much value compared to printing it with the confirmations.
Additionally we should allow the user to at least know when the XMR can be expected in the user's wallet, otherwise the swap ends like this:
```
2021-03-04 13:49:19 INFO Monero lock tx received 5 out of 5 confirmations
```
This is just not very informative - yes, the final transaction is an implementation detail, but I don't think we should hide the transactions from the user. By printing the tx-hash for spending from the lock-tx into the user wallet we ensure the user knows that the XMR can now be expected in the user wallet.
---
To add context, here the complete log (with debug enabled) **before** this change:
```
2021-03-04 13:30:46 DEBUG Database and seed will be stored in /Users/dakami/Library/Application Support/xmr-btc-swap
2021-03-04 13:30:46 DEBUG Starting monero-wallet-rpc on port 56145
2021-03-04 13:30:51 DEBUG Requesting quote
2021-03-04 13:30:51 INFO Received quote: 1 XMR = 0.00433500 BTC
2021-03-04 13:30:51 INFO Still got 0.01018746 BTC left in wallet, swapping ...
2021-03-04 13:30:51 INFO Spot price for 0.00500000 BTC is 1.153402537485 XMR
2021-03-04 13:30:52 DEBUG Starting execution setup with 12D3KooWCdMKjesXMJz1SiZ7HgotrxuqhQJbP5sgBm2BwP1cqThi
2021-03-04 13:30:55 INFO Published Bitcoin 3a6690a962191529892318819fb20e7f1ac4625400e64ee734056a9b2a17ad8f transaction as lock
2021-03-04 13:41:13 DEBUG Received Transfer Proof from 12D3KooWCdMKjesXMJz1SiZ7HgotrxuqhQJbP5sgBm2BwP1cqThi
2021-03-04 13:42:11 INFO Monero lock tx received 1 out of 5 confirmations
2021-03-04 13:45:33 INFO Monero lock tx received 2 out of 5 confirmations
2021-03-04 13:47:49 INFO Monero lock tx received 3 out of 5 confirmations
2021-03-04 13:48:56 INFO Monero lock tx received 4 out of 5 confirmations
2021-03-04 13:49:19 INFO Monero lock tx received 5 out of 5 confirmations
2021-03-04 13:49:19 DEBUG Encrypted signature sent
2021-03-04 13:49:19 DEBUG Alice acknowledged encrypted signature
2021-03-04 13:49:19 DEBUG watching for tx: e5569d3f0bcccac95252dffaebe74ead0360c09b76bc762de890aaa0e51afbcf
2021-03-04 13:49:20 DEBUG Received protocol error "missing transaction" from Electrum, retrying...
2021-03-04 13:49:22 DEBUG Received protocol error "missing transaction" from Electrum, retrying...
```
Co-authored-by: Daniel Karzel <daniel@comit.network>
Print tx-hashes for monero transactions to allow Bob to look the transaction up in block explorer.
The story of Bab:
Our famous actor Bob has a brother named Bab.
In school they were often mixed up, because their names were so similar.
Eventually Bab renamed himself into Barbara, but that was even more confusing for now he
carried a female name even though he was not female. Bob wanted to help his brother and told him he
could just go for Bub. But that did not solve anything. Fun fact: Bub is actually married to Alice.
Previously, the user neither knew the price nor the maximum quantity
they could trade. We now request a quote from the user and display
it to them.
Fixes#255.
This reduces the overall amount of LoC that imports take up in our
codebase by almost 100.
It also makes merge-conflicts less likely because there is less
grouping together of imports that may lead to layout changes which
in turn can cause merge conflicts.
265: Replace quote with spot-price protocol r=thomaseizinger a=thomaseizinger
This is essentially functionally equivalent but includes some
cleanups by removing a layer of abstraction: `spot_price::Behaviour`
is now just a type-alias for a request-response behaviour.
Co-authored-by: Thomas Eizinger <thomas@eizinger.io>
This is essentially functionally equivalent but includes some
cleanups by removing a layer of abstraction: `spot_price::Behaviour`
is now just a type-alias for a request-response behaviour.
1. We can generalize the signing interface by passing a PSBT in
instead of the `TxLock` transaction.
2. Knowing the transaction ID of a transaction that we are about
to sign is not very useful. Instead, it is much more useful to know
what failed. Hence we add a `.context` to the call of `sign_and_finalize`.
3. In case the signing succeeds, we will immediately broadcast it
afterwards. The new broadcasting interface will tell us that we broadcasted
the "lock" transaction.
We eliminate unnecessary layers of indirection for broadcasting logic
and force our callers to provide us with the `kind` of transaction
that we are publishing.
Eventually, we can replace this string with some type-system magic
we can derive the name from the actual transaction. For now, we just
require the caller to duplicate this information because it is faster
and good enough TM.
This struct is a wallet. The only thing it can meaningfully broadcast
are transactions. The fact that they have to be signed for that is
implied. You cannot broadcast unsigned transactions.
Abstracting over the individual bits of functionality of the wallet
does have its place, especially if one wants to keep a separation
of an abstract protocol library that other people can use with their
own wallets.
However, at the moment, the traits only cause unnecessary friction.
We can always add such abstraction layers again once we need them.
Log messages are ideally as close to the functionality they are talking about, otherwise we might end up repeating ourselves on several callsites or the log messages gets outdated if the behaviour changes.
If communication with the other party fails the program should stop and the user should see the respective error.
Communication errors are handled in the event-loop. Upon a communication error the event loop is stopped.
Since the event loop is only stopped upon error the Result returned from the event loop is Infallible.
If one of the two futures, event loop and swap, finishes (success/failure) the other future should be stopped as well.
We use tokio::selec! to stop either future if the other stops.
Failure does not express what the error represents. It is only used for communication
errors for quote requests, receiving the XMR transfer proof and sending the encryption signature.
These traits were only used once within the `TxLock` constructor.
Looking at the rest of the codebase, we don't really seem to follow
any abstractions here where the protocol shouldn't know about the
exact types that is being passed in.
As such, these types are just noise and might as well be removed in
favor of simplicity.
The only reason we need this argument is because we need to access
the output descriptor. We can save that one ahead of time at when
we construct the type.
For transitioning to state4 we either go into a redeem or a cancellation scenario.
The function name state4 is misleading, because it is only used for cancellation scenarios.
This TDOO is misleading, because - to our current knowledge - it is impossible for
Bob to retrieve the exact inclusion block-height of the lock transaction (send by Alice).
The wallet RPC is only capable of retrieving the inclusion block height of a transaction
through `get_payments` and `get_bulk_payments` which requires the `payment_id`.
The `payment_id` can be retrieved through `get_transfer_by_txid` which states
"Show information about a transfer to/from this address." - however the address that the
transfer goes to is not part of Bob's wallet yet! Thus, it is impossible for Bob to use
`get_transfer_by_txid` which in turn means Bob is unable to use `get_payments`.
The only possible way for Bob to know the exact inclusion block/height of the lock transaction
would be if Alice sends it over to Bob. But for that Alice would have to extract it she would have
to wait for confirmation - which she currently does not and might never do. Even if she does await
the first confirmation before sending the transfer proof the solution for retrieving the inclusion
block-height is not fleshed out on her side yet.
In order to ensure that we can atomically generate_from_keys and then reload a wallet,
we have to wrap the client of the monero wallet RPC inside a mutex.
When introducing the Mutex I noticed that several inner RPC calls were leaking to the
swap crate monero wallet. As this is a violation of boundaries I introduced the traits
`GetAddress`, `WalletBlockHeight` and `Refresh`.
Note that the monero wallet could potentially know its own public view key and
public spend key. If we refactor the wallet to include this information upon wallet
creation we can also generate addresses using `monero::Address::standard`.
To achieve this we also:
- upgrade rust-bitcoin to 0.26
- upgrade bitcoin-harness to latest version (which also depends bitcoin 0.26)
- upgrade to latest edcsa-fun
- replace cross_curve_dleq proof with sigma_fun (to avoid an upgrade dance over there)
We are aware of issues of timeouts when waiting for acknowledgements.
Also, to properly supports acks in a multiple swap context, we need to
revert to doing event processing on the behaviour so that we can link
leverage the `RequestResponse` libp2p behaviour and link the messages
requests ids to swap ids when receiving an ack or response.
Acks are usefully for specific scenarios where we queue a message on the
behaviour to be sent, save as sent in the DB but crash before the
message is actually sent. With acks we are able to resume the swap,
without ack, the swap will abort (refund).
