Commit Graph

2555 Commits

Author SHA1 Message Date
Rusty Russell
cf43e44378 common/features: don't use internal global.
Turns out that unnecessary: all callers can access the feature_set,
so make it much more like a normal primitive.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-04-03 13:13:21 +10:30
Rusty Russell
23a36ac8c8 plugins: expose onion_message hook (EXPERIMENTAL_FEATURES)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-04-02 14:32:38 +10:30
Rusty Russell
42af815932 lightningd: EXPERIMENTAL_FEATURES JSON command to sendonionmessage.
It's a fairly complex API, but maps simply onto the spec.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-04-02 14:32:38 +10:30
Rusty Russell
f7fe8a952d lightningd: forward onion messages.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-04-02 14:32:38 +10:30
Rusty Russell
dd690553b8 channeld: handle onion messages.
We do most of the decoding here, and just hand the results to lightningd.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-04-02 14:32:38 +10:30
Rusty Russell
26e906a8aa sendpay: do route parsing inside parameter parsing.
This makes "check" more accurate, and also simplifies the code a little.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-04-02 14:32:38 +10:30
Rusty Russell
b7db588a8a plugin: remove boutique features.
We now manipulate the global features directly.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-31 13:36:02 +02:00
Rusty Russell
a7cc6d33c3 plugins: actually change global features when plugins tell us to.
This cleans up the boutique handling of features, and importantly, it
means that if a plugin says to offer a feature in init, we will now
*accept* that feature.

Changelog-Fixed: Plugins: setting an 'init' feature bit allows us to accept it from peers.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-31 13:36:02 +02:00
Rusty Russell
c95e58ad4b subdaemons: initialize feature routines with explicit feature_set.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-31 13:36:02 +02:00
Rusty Russell
afb76392e4 common/features: use bitmaps internally, have explicit init function.
This is to prepare for dynamic features, including making plugins first
class citizens at setting them.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-31 13:36:02 +02:00
Rusty Russell
d881a4bd66 BOLT: update to latest version.
This is all typo/clarity fixes, no substantive changes.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-31 13:36:02 +02:00
Rusty Russell
5d4620484c lightningd: remove json_close force option.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: JSON: `close` `force` parameter removed (deprecated in 0.7.2.1)
2020-03-30 12:47:01 +02:00
Rusty Russell
7f32a844f9 listnodes/listpeers/peer_connected: remove deprecated globalfeatures and localfeatures
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: JSON: `listnodes` `globalfeatures` output (`features` since in 0.7.3).
Changelog-Removed: JSON: `listpeers` `localfeatures` and `globalfeatures` output (`features` since in 0.7.3).
Changelog-Removed: JSON: `peer_connected` hook `localfeatures` and `globalfeatures` output (`features` since in 0.7.3).
2020-03-30 12:47:01 +02:00
Rusty Russell
75838341a7 fundchannel/fundchannel_start: remove deprecated satoshi parameter
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: JSON: `fundchannel` and `fundchannel_start` `satoshi` parameter removed (renamed to `amount` in 0.7.3).
2020-03-30 12:47:01 +02:00
Rusty Russell
0007af0a4e lightningd: remove deprecated 'description' parameter.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: JSON: `sendpay` `description` parameter removed (renamed to `label` in 0.7.0).
2020-03-30 12:47:01 +02:00
darosior
610aa0b8e3 bcli: register --dev-max-fee-multiplier on our side
That way we pass the real min_acceptable (SLOW/2) and max_acceptable
(URGENT * 10) feerates to lightningd.
2020-03-30 20:17:18 +10:30
darosior
8a3295b6a8 chaintopology: Add the cost of HTLCs transactions to json_feerates
Changelog-Changed: "htlc_timeout_satoshis" and "htlc_success_satoshis" fields have been added to the `feerates` command.
2020-03-30 20:17:18 +10:30
darosior
d4fe4073a4 lightning/bitcoind: adapt and batch fees estimations
This adapts our fee estimations requests to the Bitcoin backend to the
new semantic, and batch the requests.

This makes our request for fees much simpler, and leaves some more
flexibility for a plugin to do something smart (it could still lie before
but now it's explicit, at least.) as we don't explicitly request
estimation for a specific mode and a target.

Changelog-Changed: We now batch the requests for fee estimation to our Bitcoin backend.
Changelog-Changed: We now get more fine-grained fee estimation from our Bitcoin backend.
2020-03-30 20:17:18 +10:30
darosior
dce2e87928 chaintopology: better feerate targets differentiation
We kept track of an URGENT, a NORMAL, and a SLOW feerate. They were used
for opening (NORMAL), mutual (NORMAL), UNILATERAL (URGENT) transactions
as well as minimum and maximum estimations, and onchain resolution.

We now keep track of more fine-grained feerates:
- `opening` used for funding and also misc transactions
- `mutual_close` used for the mutual close transaction
- `unilateral_close` used for unilateral close (commitment transactions)
- `delayed_to_us` used for resolving our output from our unilateral close
- `htlc_resolution` used for resolving onchain HTLCs
- `penalty` used for resolving revoked transactions

We don't modify our requests to our Bitcoin backend, as the next commit
will batch them !

Changelog-deprecated: The "urgent", "slow", and "normal" field of the `feerates` command are now deprecated.
Changelog-added: The fields "opening", "mutual_close", "unilateral_close", "delayed_to_us", "htlc_resolution" and "penalty" have been added to the `feerates` command.
2020-03-30 20:17:18 +10:30
darosior
ad4bcfde53 chaintopology: add delayed_to_us, htlc, and penalty feerates 2020-03-30 20:17:18 +10:30
darosior
8cbc0038bc onchaind: a feerate per transaction type
This allows us to set more fine-grained feerate for onchain resolution.

We still give it the same feerate for all types, but this will change as
we move feerates to bcli.
2020-03-30 20:17:18 +10:30
Michael Schmoock
1aa8c90c73 chore: refactor and unify LOCAL/REMOTE helpers for receivable_msat 2020-03-27 12:33:23 +01:00
Michael Schmoock
ed56b22094 fix: calc commit_txfee_spend/recv for spendable or receivable 2020-03-27 12:33:23 +01:00
Michael Schmoock
33c8afea80 feat: adds receivable_msat to listpeers
Changelog-Added: JSON API: `listpeers` now has `receivable_msat`
2020-03-27 12:33:23 +01:00
Rusty Russell
e1ba42f139 lightningd: always use BROKEN log level if we can't spawn a daemon.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-26 13:29:05 +01:00
Rusty Russell
3e311b2510 lightningd: fix assert() if spawning openingd fails.
My node crashed as follows:

	lightningd: lightningd/peer_control.c:957: peer_connected: Assertion `!peer->uncommitted_channel' failed.

In the logs I found:

	Running lightning_openingd: Cannot allocate memory

Which reveals that we're not freeing uc in that path!

Changelog-Fixed: Fix assertion on reconnect if we fail to run openingd.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-26 13:29:05 +01:00
Rusty Russell
8c984fbf1d common: make sphinx.c use hmac.c.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-25 14:26:28 +10:30
lisa neigut
563b46814b plugins: remove 'blank' option parsing for bool
bools must be set!
2020-03-21 16:29:52 +10:30
lisa neigut
42cce55b45 plugins: add 'flag' type for plugin options
Updates the plugin docs to include more detailed info about how options
work.

Changelog-Added: Plugins: 'flag'-type option now available.
2020-03-21 16:29:52 +10:30
Michael Schmoock
af7e879308 fix: rfc #740 requires 100% feespike margin
Changelog-Fixed: Use lightning-rfc #740 feespike margin factor of 2
2020-03-19 18:56:12 +01:00
Rusty Russell
24984ec680 common/sphinx: add realm flag so we can avoid legacy parsing.
For messages, we use the onion but payload lengths 0 and 1 aren't special.
Create a flag to disable that logic.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-17 18:47:52 +01:00
darosior
06e9a9f31f bitcoind: check that Bitcoin plugin is alive before requesting 2020-03-11 16:01:36 -05:00
lisa neigut
41b18050a0 plugin: add in deprecated_api behavior and test
we also check that the node isn't running now, for extra pedancity
2020-03-10 13:25:36 +10:30
lisa neigut
cac5a0cd1d plugins: use stricter parsing for option values
also: convert the stored int value from 'int' to 's64'

atoi fails silently, returning a zero. instead we use the more robust
strtoll which will allow us fail with an error.

we also make the parsing for bools stricter, only allowing plausibly
boolean values to parse.
2020-03-10 13:25:36 +10:30
lisa neigut
4e30a82f09 plugins: pass back opts as indicated type. fixes #3577
Changelog-Fixed: Plugins: if an option has a type int or bool, return the option as that type to the plugin's init
2020-03-10 13:25:36 +10:30
Christian Decker
d2688bbaf5 plugin: Unwrap the rpc_command payload
We were nesting like the following:

```json
{"params": {
  "rpc_command": {
    "rpc_command": {
    }
  }
}
```

This is really excessive, so we unwrap once, and now have the following:

```json
{"params": {
  "rpc_command": {
  }
}
```

Still more wrapping than necessary (the method is repeated in the `params`
object), but it's getting closer.

Changelog-Deprecated: JSON-RPC: Removed double wrapping of `rpc_command` payload in `rpc_command` JSON field.

Suggested-by: @fiatjaf
Signed-off-by: Christian Decker <@cdecker>
2020-03-10 11:42:58 +10:30
Vasil Dimov
66eca2ff3f lightningd: always broadcast the latest close tx
Before this patch we would only update `channel->last_tx` with the newly
proposed closure tx from the peer if the fee of the new one was lower.

In negotiations where we are at the higher end and the peer starts
lower, all peer's subsequent proposals will be higher than his initial
proposal and in this case we would never update `channel->last_tx`
and would wrongly broadcast his initial proposal at the end of the
negotiation.

Fixes https://github.com/ElementsProject/lightning/issues/3549

Changelog-Fixed: Always broadcast the latest close transaction at the end of the close fee negotiation, instead of sometimes broadcasting the peer's initial closing proposal.
2020-03-10 11:40:19 +10:30
Rusty Russell
c92e782e22 wire: add fromwire_tal_arrn() helper.
Does the allocation and copying; this is useful because we can
avoid being fooled into doing giant allocations.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-03-09 16:04:56 +01:00
darosior
1fd45a061b bitcoind: timeout if the Bitcoin plugin never completes the handshake
Reported-by: Vasil Dimov <@vasild>
2020-03-05 15:06:38 -06:00
Christian Decker
ac5cba2243 htlc: Return an invalid payload error if payload is invalid
We somehow were always returning `badonion` when really it is an invalid
payload inside the onion.
2020-03-04 22:57:24 +01:00
Christian Decker
959687bf6d onion: Pass the position and type of the failing TLV type out
We'll need this when returning an error to the sender.
2020-03-04 22:57:24 +01:00
Michael Schmoock
6c50185865 fix: cppcheck ignore two false positives on uninitvar 2020-03-04 14:04:51 +10:30
ZmnSCPxj jxPCSnmZ
d9b2482415 lightningd/hsm_control.c: Implement getsharedsecret.
ChangeLog-Added: New `getsharedsecret` command, which lets you compute a shared secret with this node knowing only a public point. This implements the BOLT standard of hashing the ECDH point, and is incompatible with ECIES.
2020-02-28 14:45:50 +10:30
Rusty Russell
f8a21f16c9 lightingd: do a local short_channel_id lookup for forwarding.
Even without optimization, it's faster to walk all the channels than
ping another daemon and wait for the response.

Changelog-Changed: Forwarding messages is now much faster (less inter-daemon traffic)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-02-28 09:44:47 +10:30
Rusty Russell
40e3566e9a lightningd: use the async mechanism for channel_update access.
Instead of saving a stripped_update, we use the new
local_fail_in_htlc_needs_update.

One minor change: we return the more correct
towire_temporary_channel_failure when the node is still syncing.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-02-28 09:44:47 +10:30
Rusty Russell
247d249ea8 gossipd: provide helper to get a channels cupdate, create routine to use it.
The idea is that gossipd can give us the cupdate we need for an error, and
we wire things up so that we ask for it (async) just before we send the
error to the subdaemon.

I tried many other things, but they were all too high-risk.

1. We need to ask gossipd every time, since it produces these lazily
   (in particular, it doesn't actually generate an offline update unless
   the channel is used).
2. We can't do async calls in random places, since we'll end up with
   an HTLC in limbo.  What if another path tries to fail it at the same time?
3. This allows us to use a temporary_node_failure error, and upgrade it
   when gossipd replies.  This doesn't change any existing assumptions.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-02-28 09:44:47 +10:30
Rusty Russell
2aad3ffcf8 common: tal_dup_talarr() helper.
This is a common thing to do, so create a macro.

Unfortunately, it still needs the type arg, because the paramter may
be const, and the return cannot be, and C doesn't have a general
"(-const)" cast.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-02-27 14:16:16 +10:30
Rusty Russell
684ed4231f common/wireaddr: don't include lightningd/lightningd.
common should not include specific per-daemon files.  Turns out this
caused a lot of indirect includes to be exposed.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-02-27 14:16:16 +10:30
Christian Decker
8f87579589 cleanup: Remove current_plugin from plugin_hook_request
It was a pointer into the list of plugins for the hook, but it was rather
unstable: if a plugin exits after handling the event we could end up skipping
a later plugin. We now rely on the much more stable `call_chain` list, so we
can clean up that useless field.
2020-02-27 09:21:44 +10:30
Christian Decker
23149c3daa plugin: Actually wait the 20 seconds promised in the docs
We promised we'd be waiting up to 20 seconds, but were only waiting for
10. Fix that by bumping to the documented 20.
2020-02-27 09:21:44 +10:30