Split `enforceTor` into `tor.proxy` and `tor.enforce`.
By enabling `tor.proxy` without `tor.enforce`, a service can accept
incoming clearnet connections.
E.g., this allows setting up a Tor-proxied bitcoind node that accepts
RPC connections from LAN.
When NixOS is already running and Tor is restarted due to config
changes, `/var/lib/tor/state` may be present even when Tor has not
yet finished setting up onion services.
This caused the previous version of `onion-addresses` to not wait for
Tor and to skip not yet present onion service files.
`onion-addresses` now waits until each required onion service file
has appeared.
4a74b7de08 clightning: work around unsupported seccomp syscall (Erik Arvstedt)
38a843d005 clightning: update python pkgs to new version (Erik Arvstedt)
6ad7107ddb update nixpkgs (Erik Arvstedt)
f58d67677e netns-isolation: separate host and netns setup (Erik Arvstedt)
cb6e5ef702 netns-isolation: fix routing issues due to netns restarting (Erik Arvstedt)
7f77147b60 makeShell: minor improvements (Erik Arvstedt)
a5730eb736 makeShell: make the help msg a shell derivation variable (Erik Arvstedt)
Pull request description:
ACKs for top commit:
jonasnick:
ACK 4a74b7de08
Tree-SHA512: 75454b51db6d7ab41590d8579e0a5136e5ac1be78d5c2f547c6ef1982c0de679968879bb9bac57dd66413f59a4659236601ab75414486b0137c7c43d73d22759
- Improves readability
- `netns exec ...` (called via `netnsIptables`) incurs a large
overhead: In addition to netns switching, a mount namespace
is setup and populated with the contents of /etc/netns/<ns>/.
Instead, simply run `nsenter`.
Previously, restarting a service implied restarting its netns.
For unknown reasons, this sometimes caused the netns-local address
to not be routable from the root netns for up to 20 s.
I.e., the service was sometimes unreachable after restarting.
Now the netns is no longer stopped when the service is stopped.
Otherwise liquidd startup fails. This bug was not covered by our tests,
because we're not combining `regtest` with `secure-node`.
But nixbitcoin.org does, which should suffice for now.
This simplifies the host's address configuration.
This also removes unused addresses that are returned when resolving
container hostnames via nss-mymachines:
`getent ahosts nb-test`
There is no security reason why pruning should be enabled and therefore it
surprises users. Turning on pruning in the first place was simply a mistake.
- README: add matrix room
- examples/configuration.nix: explain why bitcoind is enabled by default
- btcpayserver: group lnd service settings
- clightning:
Use public onion port only when the onion service is public
This allows users to enable the onion service while announcing a
non-onion public address.
- netns-isolation: move `readOnly` attr to the top
- tests: use mkDefault to allow for easier overriding
- tests/btcpayserver: test web server response
This allows whitelisting local services without implicitly
whitelisting all inbound onion connections, which would happen when
setting bitcoind/liquidd option `whitelist=localhost`.
Used by electrs and nbxplorer, which requires the unsafe `mempool`
permission.
This removes the module-level dependency from onion-services to
bitcoind.
Due to the `or false` fallback, there's no dependency added in
the reverse direction.
In particular, this allows us to not add a dependency on liquidd in
the following commit.
Whitelisting localhost implicitly whitelists all inbound onion
connections. This prevents banning misbehaving inbound onion peers
and enables message `mempool` which can cause privacy leaks.
Instead, grant `download` as the single bitcoind whitelist permission, which
should be safe for onion peers.
Remove liquidd whitelisting because it doesn't support fine-grained permissions.
After a cursory glance at the nbxplorer code I think that nbxplorer
requires none of the other default whitelist permissions (noban, mempool,
relay).
Details: https://github.com/dgarage/NBXplorer/issues/344
This re-enables onion tagging while still supporting untagged connections.
Onion sockets are not yet supported in the latest liquidd/elements
version 0.18.1.12 available on nixpkgs.
3781a85c9b joinmarket: enable Agora as a third IRC server (nixbitcoin)
ced1637d07 joinmarket: share IRC server definitions between jm and ob-watcher (Erik Arvstedt)
59fc003ebd joinmarket: 0.9.1 -> 0.9.2 (nixbitcoin)
Pull request description:
ACKs for top commit:
erikarvstedt:
ACK 3781a85c9b
Tree-SHA512: 5ec919d2291ecf96fb4ca880f3dbeabff13f2bab71822db893ebbaba1b95463666b098ccc1412a1b56f327a231e10c1f2d47feb0f520fce349ab243d398bf7b4
- `waitfornewblock` was previously not included in the public RPC
whitelist because it's reserved for testing and marked as hidden
in bitcoind.
- electrs changed its verbosity settings. `-vv` is now the best choice
for normal usage.
- bitcoind option `dataDirReadableByGroup` is now unused.
Because it can be valuable for other use cases and implementing
it is intricate, we're keeping it for now.
- test: keep `nc` connection open because otherwise the electrs
RPC server would now close the connection before sending a response.
Previously, Tor was always enabled because `cfg` was always nonempty
(via definitions at `Set sensible defaults for some services`).
Now only enable Tor if there are active onion services.
Also rename var `services` -> `onionServices` to improve readability in
section `Set getPublicAddressCmd ...` where the same name is also used for
option `config.services`.
Previously, the glob (*) returned '*' when no files existed in the
secrets dir, leading to error `chown: cannot access '*'`.
Now `unprocessedFiles` is empty when there are no secrets.
Also remove the unneeded sorting of `unprocessedFiles` and
remove redundant leading zero in the default mode.
- `discover` is automatically disabled by bitcoind because we're
setting `externalip` via the `nix-bitcoin.onionServices` mechanism
- `bech32` is bitcoind's default addresstype
