Commit Graph

12076 Commits

Author SHA1 Message Date
Rusty Russell 4e39b3ff3d hsmd: don't use point32 for bolt12, but use pubkeys (though still always 02)
This is the one place where we hand point32 over the wire internally, so
remove it.

This is also our first hsm version change!

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-10-26 11:29:06 +10:30
Rusty Russell bed905a394 lightningd: use 33 byte pubkeys internally.
We still use 32 bytes on the wire, but internally don't use x-only.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-10-26 11:29:06 +10:30
Rusty Russell 987adb9718 Makefile: check that hsm_version.h changes if wire/hsmd_wire.csv contents does
Otherwise I know we'll miss it.  Simply check for a mention: we could well
change things multiple times within a single release.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-10-26 11:29:06 +10:30
Rusty Russell fee9a7ce04 hsmd: introduce a simple API versioning scheme.
With the rise of external HSMs like VLS, this is no longer an
internal-only API.  Fortunately, it doesn't change very fast so
maintenance should not be a huge burden.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-10-26 11:29:06 +10:30
niftynei e00857827f funder: cleanup datastore on state-change/channel failure
Let's not leave old state hanging around! Note that this fires
for pretty much every/any channel (even if we're not the opener).
2022-10-20 13:42:41 +02:00
niftynei efd096dc96 funder: filter prev-outs such that we only use still unspent ones
If for some reason a utxo we used previously is no longer 'unspent',
we shouldn't use it for the next transaction.
2022-10-20 13:42:41 +02:00
niftynei 38e2428f12 funder: use utxopsbt to build psbt for RBFs
We use the saved previous outputs (plus maybe some new ones?) to build a
psbt for an RBF request.

RBFs utxo reuse is now working so we can unfail the test (and update
it to reflect that the lease sticks around through an RBF cycle).

Changelog-Fixed: Plugins: `funder` now honors lease requests across RBFs
2022-10-20 13:42:41 +02:00
niftynei 45acc20a8d funder: use previous outputs in count towards available funding
Still need to use them to build the PSBT for the rbf however.
2022-10-20 13:42:41 +02:00
niftynei 7733c2b0f4 funder: pull out previous input list from datastore on RBF
It'd be nice to know which utxos we used previously, so we can rebuild a
transaction using them!
2022-10-20 13:42:41 +02:00
niftynei a5e9035e2e funder: save utxos of signed txs to datastore
We're going to need to know what utxos we used if we RBF this channel;
so we serialize our inputs and save them to the datastore under the channel_id.
2022-10-20 13:42:41 +02:00
niftynei dc9e79c445 funder: rm quote that makes nifty cringe every time she sees it 2022-10-20 13:42:41 +02:00
niftynei 85039d4f4e df: pass lease data back to funder for rbfs
let's let RBFs know about our lease info!
2022-10-20 13:42:41 +02:00
niftynei d3066ab7f9 openchannel2: may re-use rates
If more than one plugin calls `openchannel2`, the payload will get
re-freshed/re-parsed.
2022-10-20 13:42:41 +02:00
niftynei 00d3e3e492 df: for rbfs, since we know what they asked for, we can abort
if they request less than we wanted/accepted

FIXME: add a test for this?
2022-10-20 13:42:41 +02:00
niftynei fa987f2344 df: put requested_lease onto state, so it persists
We're gonna need it for rbf requests/re-negotiations
2022-10-20 13:42:41 +02:00
niftynei 8d864d2299 funder: we always pass in channel_max, no need to special case it
Parse the channel_max along with everything else.
2022-10-20 13:42:41 +02:00
niftynei 250b190e5f chainparams/dual-open: set max_supply; use for max on wumbo channels
We were leaving out the `channel_max_msat` for `openchannel2` when
channels are 'wumbo' (the conversion to msat in the json helper
overflowed, which resulted in the field not being printed)

Changelog-Changed: Plugins: `openchannel2` now always includes the `channel_max_msat`
2022-10-20 13:42:41 +02:00
Chris Guida 7c7c4c4cb3 zlib 1.2.12 yanked, update to 1.2.13 in Dockerfile 2022-10-20 13:27:38 +02:00
Rusty Russell d2633d3e6d pytest: fix flake in test_emergencyrecover
Make sure bitcoind sees tx before we mine blocks!

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-10-20 12:24:51 +02:00
niftynei 49ed0a4b9e psbt: wipe global tx scriptSig/witness data after saved to PSBT
The global tx should be "free from sin" (no scriptSig data, no witness
stacks).
2022-10-18 12:46:45 -05:00
niftynei 1c5edc14a5 bitcoin: add test for to/from wiring a bitcoin tx w/ scriptsig data
This should go to/from wire successfully, but it fails currently.

Reported-By: @ddustin
2022-10-18 12:46:45 -05:00
Christian Decker 62bfed9a8d docker: Add bitcoin-cli to the alpine dockerfile 2022-10-15 13:42:58 +02:00
Christian Decker 6fb653ef97 docker: Separate builder from runner stage in alpine docker image
Should result in smaller images.
2022-10-15 13:42:58 +02:00
Christian Decker 0acdc911a5 docker: Build the alpine docker image from a clone
This was causing some issues because it was picking up pre-built
artefacts from the host machine. By cloning first we ensure it matches
the latest commit and compiles from scratch.
2022-10-15 13:42:58 +02:00
Christian Decker 1f5e579f1b docker: Clean up dependencies for alpine build 2022-10-15 13:42:58 +02:00
briancolecoinmetrics f0d81f46f0 Clean up Dockerfile.alpine
bump Dockerfile.alpine from Alpine 3.14 to 3.16 (current stable) and remove apparently redundant dependencies.
2022-10-15 13:42:58 +02:00
briancolecoinmetrics 74652f7cf4 move alpine build dependencies to virtual package
move alpine build dependencies to virtual package so it can be easily removed all at once
2022-10-15 13:42:58 +02:00
briancolecoinmetrics 2136d5912f update package dependencies for Alpine Linux
Alpine no longer has a `python` (2) package, which is fine because it doesn't seem to be needed. Also, the listed commands didn't result in all needed dependencies being installed for runtime, so I've added that in an additional step.
2022-10-15 13:42:58 +02:00
Christian Decker a99509db36 py: Update protobuf dependency to silence dependabot 2022-10-14 19:31:39 +02:00
Rusty Russell e855ac2f9e keysend: just strip even unknown fields.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: Plugins: `keysend` now removes unknown even (technically illegal!) fields, to try to accept more payments.
2022-10-04 17:48:08 +02:00
adi2011 e7e7a7186f tests/test_misc.py: Check if funds are getting recovered on reconnecting... Changelog-None: Increasing test scope 2022-10-01 14:01:19 +02:00
okjodom 335f52d1a8 cln-rpc: implement from Secret to slice conversion 2022-09-30 10:18:12 +02:00
Dustin Dettmer 6aca9f665b devtools: Make fund_nodes compatible w/zsh
Mac is updating to using zsh in general.

The “for i in” reads strings with spaces as a single entry instead of multiple entries as sh did.

Using “while read” … “<<< $var” makes it treat each space as a new entry.

Changelog-None
2022-09-30 10:17:46 +02:00
joemphilips 52be59587c msggen: generate deprecated fields in rust.py 2022-09-29 11:18:06 +02:00
Rusty Russell 22c8cfc374 channeld: remove onion objects.
We don't actually process onion messages here any more (they moved to
connectd), but the flag and object files were still linked.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 2749b1f61e common/onion: remove old blinded payment handling.
It's been obsoleted and needs replacing; less confusing if we remove
it first.

Also, these fields are now present even without an expermintal build
(we'll control at runtime).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell c9d42d8279 bitcoin: add routine to check a Schnorr sig given a 33-byte pubkey.
There's no secp256k1 routine to do this, but we're going to need it.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 0195b41461 pytest: test that we don't change our payer_key calculation.
If we do, an upgrade would mean we can no longer get refunds on old
invoices.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 125b17b7fc devtools: enhance bolt12-cli to convert to/from hex
This makes it easier to add/delete fields manually to a bolt12 encoding.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 2fbe0f59f1 plugins/fetchinvoice: remove obsolete string-based API.
Generate the payload in the callers.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 41ef85318d onionmessages: remove obsolete onion message parsing.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 342e330b56 doc: update references to old BOLTs repo.
This reveals that common/test/run-bolt12_merkle-json.c was broken!

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Rusty Russell 49fe1c8ed7 lightningd: have `makesecret` take `hex` or `string` (just like `datastore`)
Changelog-Added: JSON-RPC: `makesecret` can take a string argument instead of hex.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-29 16:10:57 +09:30
Christian Decker 6adb1e0b4b pytest: Bypass schema verification for some RPC calls
The goal here is to test the node validation, not whether we can
trigger the schema validation with bogus values. So we bypass the
verifying RPC wrapper.
2022-09-28 15:13:07 +02:00
Christian Decker 93b315756f schema: Add `enforcedelay` to `setchannel`
This was missing in the last version
2022-09-28 15:13:07 +02:00
Justin Litchfield 51e2433087 Setchannel request is provided 2022-09-28 15:13:07 +02:00
Matt Whitlock 6324980484 channeld: do not enforce max_accepted_htlcs on LOCAL in channel reinit
If a channel goes offline while the count of outstanding outgoing HTLCs
exceeds the limit that we enforce against the peer, then the channel
could never be brought online again because `add_htlc` called by
`channel_force_htlcs` in `channeld/full_channel.c` would return
`CHANNEL_ERR_TOO_MANY_HTLCS`. The protocol specification actually does
allow us to exceed the limits that we are enforcing against the peer;
we are only prohibited from exceeding the limits that the peer is
enforcing against us. `add_htlc` takes an `enforce_aggregate_limits`
parameter that appears to have been intended for `channel_force_htlcs`
to exempt the local node from obeying the limits that it is enforcing
against the peer, but this parameter was only being respected for the
total HTLC value-in-flight check but not for the HTLC count check. This
commit respects the parameter for the HTLC count check as well and
resolves the problem of "Could not restore HTLCs".

Fixes: #5636
Changelog-Fixed: channeld: Channel reinitialization no longer fails when the number of outstanding outgoing HTLCs exceeds `max_accepted_htlcs`.
2022-09-28 14:00:20 +02:00
Rusty Russell f00cc23f67 sphinx: rename confusing functions, ensure valid payloads.
"sphinx_add_hop" takes a literal hop to include,
"sphinx_add_modern_hop" prepends the length.  Now we always prepend a
length, make it clear that the literal version is a shortcut:

* sphinx_add_hop -> sphinx_add_hop_has_length
* sphinx_add_modern_hop -> sphinx_add_hop

In addition, we check that length is actually correct!  This means
`createonion` can no longer create legacy or otherwise-invalid onions:
fix tests and update man page to remove legacy usage.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: JSON-RPC: `createonion` no longer allows non-TLV-style payloads.
2022-09-28 13:40:57 +02:00
Rusty Russell 8771c86379 common/onion: expunge all trace of different onion styles.
In particular, remove special routines to pull length: it's there,
take it and check it yourself.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2022-09-28 13:40:57 +02:00
Rusty Russell c8ad9e18a9 common/onion: remove all trace of legacy parsing.
We still have an "enum forward_style" for the database, where old-style
forwards can still exist.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: Protocol: we no longer forward HTLCs with legacy onions.
2022-09-28 13:40:57 +02:00