615: Update default Monero node URLs and info r=thomaseizinger a=da-kami
xmr.to nodes are not served at the xmr.to domain but moved to melo.tools
Co-authored-by: Daniel Karzel <daniel@comit.network>
The Rust compiler doesn't guarantee in which order field initialization
are executed. By extracting them, we can make sure they run in a certain
order. This will be important as we add more validations that can
fail.
For builds made directly on the tag, the output of `--version` will
not change. For builds not made on a tagged commit, the output will
look something like this:
```
> swap --version
swap 0.7.0-117-g93161f9
```
Fixes#409.
We emit an `info!` log for every peer that we discover but only ever
emitted a `debug!` log if we fail to connect. This leads to a situation
where the user would run `swap list-sellers`, the logs would say
"Discovered XYZ at ABC" but then get a potentially empty table.
To not confuse the user, we include unreachable nodes in the table output.
For example:
```
Connected to rendezvous point, discovering nodes in 'xmr-btc-swap-testnet' namespace ...
Discovered peer 12D3KooWPZ69DRp4wbGB3wJsxxsg1XW1EVZ2evtVwcARCF3a1nrx at /dns4/ac4hgzmsmekwekjbdl77brufqqbylddugzze4tel6qsnlympgmr46iid.onion/tcp/8765
+-------+--------------+--------------+-------------+----------------------------------------------------------------------------------------------------------------------------------------+
| PRICE | MIN_QUANTITY | MAX_QUANTITY | STATUS | ADDRESS |
+============================================================================================================================================================================================+
| ??? | ??? | ??? | Unreachable | /dns4/ac4hgzmsmekwekjbdl77brufqqbylddugzze4tel6qsnlympgmr46iid.onion/tcp/8765/p2p/12D3KooWPZ69DRp4wbGB3wJsxxsg1XW1EVZ2evtVwcARCF3a1nrx |
+-------+--------------+--------------+-------------+----------------------------------------------------------------------------------------------------------------------------------------+
```
This command uses a rendezvous node to find sellers (i.e. ASBs) and query them for quotes.
Sellers, that can be dialed and queried for a quote will be listed.
Some network and application specific code does not belong in the protocol module and was moved.
Eventloop, recovery and the outside behaviour were moved to the respective application module because they are application specific.
The `swap_setup` was moved into the network module because upon change both sides will have to be changed and should thus stay close together.
Instead of splitting up the transports into capabilities, we compose
them directly for each application. This allows us to remove the
websocket transport for the CLI which is really only needed for the
ASB to allow retrieval of quotes via the browser.
It appears to be more stable.
Encountered issues with the previous setup, `monero-wallet-rpc` logs:
```
2021-05-24 04:23:54.852 E !r. THROW EXCEPTION: tools::error::no_connection_to_daemon
2021-05-24 04:23:54.857 E Exception at while refreshing, what=no connection to daemon
```
This includes testing CLI commandline args
Clap's `default_value_with` actually did not work on `Subcommand`s because the parent's flags were not picked up.
This was fixed by changing parameters dependent on testnet/mainnet to options.
This problem should have been detected by tests, that's why the command line parameter tests were finally (re-)added.
Thanks to @rishflab for some pre-work for this.
In order to allow people to plug into public nodes / be more flexible with their own setup we now enforce specifying the monero daemon port to be used by the `monero-wallet-rpc`.
In the past we had problems with flags/parameter changes several times, where on instance was changed, buy another one was missed. This should mitigate this problem.
This patch introduces structs for all duplicated parameters and uses flatten to only have one point for changes.
Additionally removes all mentions of `alice` from the commands / variables. This code is on an application level and should not be concerned with swap protocol roles.
Using the same default directory as data-/config-dir has caused unwanted side effects when running both applications on the same machine.
Use these directory names:
- ASB: xmr-btc-swap/asb
- CLI: xmr-btc-swap/cli
Since the functionality is now application specific the respective functions were moved into the appropriate module of the application.
Using the same default directory as data-/config-dir has caused unwanted side effects when running both applications on the same machine.
Use these directory names:
- ASB: xmr-btc-swap-asb
- CLI: xmr-btc-swap-cli
Since the functionality is now application specific the respective functions were moved into the appropriate module of the application.
This PR does a few things.
* It adds a TorTransport which either dials through Tor's socks5 proxy or via clearnet.
* It enables ASB to register hidden services for each network it is listening on. We assume that we only care about different ports and re-use the same onion-address for all of them. The ASB requires to have access to Tor's control port.
* It adds support to dial through a local Tor socks5 proxy. We assume that Tor is always available on localhost. Swap cli only requires Tor to be running so that it can send messages via Tor's socks5 proxy.
* It adds a new e2e test which swaps through Tor. For this we assume that Tor is currently running on localhost. All other tests are running via clear net.
EnvFilter is applied globally. This means you cannot log at INFO level
to the terminal and at DEBUG level to log files. To get a around this
limitation I had to implement the layer trait on a new type and filter
in the on_event() trait method. Each swap has its own log file denoted
by its swap_id. The logger appends to the existing file when resuming a
swap.
Closes#278
This allows loading the seller-peer-id from the database upon resuming a swap.
Thus, the parameters `--seller-peer-id` is removed for the `resume` command.
Other than the peer-id the multi address of a seller can change and thus is
still a parameter. This parameter might become optional once we add DHT support.
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>
The CLI has sensible default values for all parameters,
thus a config file is not really an advantage but just
keeps getting in our way, so re remove it.
The type hints are generated from the field names. This has the
unfortunate consequence of the config field becoming file_path which
does not really make sense people working on the codebase.
Thomas Eizinger
bors[bot]
Daniel Karzel
Philipp Hoenisch
rishflab