This means only a single gossmap, and they already share the fetchinvoice-noconnect option
and autoconnect code.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: Plugins: the `fetchinvoice` plugin has been combined into the `offers` plugin.
Making sure that we are able to fetch the invoice
from an offer without description
E pyln.client.lightning.RpcError: RPC call failed: method:
offer, payload: {'amount': '2msat'}, error: {'code': -32602, 'message':
'bolt12: Offer does not contain a description: invalid token
\'"lno1qgsqvgnwgcg35z6ee2h3yczraddm72xrfua9uve2rlrm9deu7xyfzrcgqyppvggz953rvg9rtxj8lalh43z8epwydjfrmffn3y3p5qz5cywpu09rr4vs"\''}
Link: https://github.com/ElementsProject/lightning/issues/7405
Link: https://github.com/ElementsProject/lightning/issues/7404
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
Changelog-Added: Protocol: onion messages are now supported by default.
Changelog-Deprecated: Config: the --experimental-onion-messages option is ignored (on by default).
However fast we can handle them, it's antisocial to allow others to
make us spam the rest of the network.
Changelog-Protocol: onion messages: we limit incoming to 4 per second, allowing a little burst.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
A bit tricky, since we get more than one message at a time. However,
this just means we go over quota for a bit, and will get caught when
those are sent (we do this for a single message already, so it's not
that much worse).
Note: this not only limits sending, but it limits the actuall query
processing, which is nice.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We currently stream gossip as fast as we can, even if they start at
timestamp 0. Instead, use a simple token bucket filter and only let
them have 1MB per second (500 bytes per second for testing).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Protocol: connectd: we now throttle outgoing gossip at 1MB/second per peer.
Document and enforce the --experimental-anchors deprecation, which was somehow missed in v24.02
Changelog-Deprecated: Config: the --experimental-anchors option is ignored (on by default since v24.02).
We don't use the payment_secret in bolt12, but in onion_decode() we
do put the path_secret (if of correct length) into payment_secret. Not
realizing this confused me, so document that, and make sure we insist
on it being present for bolt12.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Shahana decided this was the optimal UX path, though I insisted that we still
report the actual problem directly when in dev mode, as a compromise.
Suggested-by: https://github.com/Amperstrand
Changelog-Changed: JSON-RPC: Do not return the contents of invalid parameters in error messages, refer to logs (use 'check' to get full error messages)
Fixes: https://github.com/ElementsProject/lightning/issues/7338
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
```
2024-06-24T05:09:32.5233603Z assert l1.rpc.listfunds()["channels"][0]["state"] == "ONCHAIN"
2024-06-24T05:09:32.5234187Z > assert l2.rpc.listfunds()["channels"][0]["state"] == "ONCHAIN"
2024-06-24T05:09:32.5234917Z E AssertionError: assert 'FUNDING_SPEND_SEEN' == 'ONCHAIN'
2024-06-24T05:09:32.5235464Z E - ONCHAIN
2024-06-24T05:09:32.5235773Z E + FUNDING_SPEND_SEEN
2024-06-24T05:09:32.5236096Z
2024-06-24T05:09:32.5236242Z tests/test_misc.py:2907: AssertionError
```
It's possible that l2 hasn't seen the onchain tx yet. What we really want to do is
wait for it, then test that l1 really can spend the output it has recovered.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
```
2024-06-24T05:14:14.9939033Z > l1.rpc.call('dev-feerate', [l2.info['id'], rate - 5])
2024-06-24T05:14:14.9939354Z
2024-06-24T05:14:14.9939466Z tests/test_connection.py:3439:
...
2024-06-24T05:14:14.9967617Z > raise RpcError(method, payload, resp['error'])
2024-06-24T05:14:14.9968833Z E pyln.client.lightning.RpcError: RPC call failed: method: dev-feerate, payload: ['022d223620a359a47ff7f7ac447c85c46c923da53389221a0054c11c1e3ca31d59', 2290], error: {'code': -1, 'message': 'Peer bad state'}
```
The disconnect can actually take a while: wait for both sides to
believe they're reconnected. Also wait a little to make sure the
feerate change doesn't create a bad signature.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
lightningd-3 penalizes lightningd-2 but then it can see the preimage for the HTLC which
has already been timed out:
```
2024-06-24T02:41:29.4633900Z lightningd-3 2024-06-24T02:33:54.073Z **BROKEN** 022d223620a359a47ff7f7ac447c85c46c923da53389221a0054c11c1e3ca31d59-onchaind-chan#1: HTLC already resolved by THEIR_HTLC_TIMEOUT_TO_THEM when we found preimage
```
This is fair: the test deliberately takes l3 offline for long enough
that the HTLC can get timed out.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I was looking into using the `threading.Condition` but since we're
already rather heavily using callbacks, this allows us to stay
single-threaded, and not having to completely hook the `setconfig`
function.
Changelog-Added: pyln-client: Added a notification mechanism for config changes
We didn't actually *change* the value you'd see, when we got a setconfig call!
Changelog-Added: pyln-client: implement setconfig hook for plugins so you can see changes in `dynamic` options.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is a difficult transition for us: this string appears in channel
types. We make the transition now in the understanding that it will
be more difficult in future.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Deprecated: JSON-RPC: `listpeers` `features` array string "option_anchors_zero_fee_htlc_tx": use "option_anchors" (spec renamed it).
Changelog-Added: JSON-RPC: `listpeers` `features` array string uses "option_anchors" for feature 22/23, following renaming in BOLT 9.
Changelog-Changed: JSON-RPC: `listclosedchannels`, `listpeerchannels`, `openchannel_update`, `openchannel_init`, `fundchannel`, `fundchannel_start` and `multifundchannel`: `channel_type` array `names` now contains "anchors" instead of "anchors_zero_fee_htlc_tx".
Changelog-Changed: lightningd: `--list-features-only` now lists "option_anchors" instead of "option_anchors_zero_fee_htlc_tx".
These were removed from the spec.
We still support existing ones, though we were the only implementation
which ever did, and only in experimental mode, so we should be able to
upgrade them and avoid a forced close, with a bit of engineering...
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We still support *existing* channels. Just not new ones (before they could,
in theory, explicitly ask for one).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This means we do have to set the network correctly though,
and also we can get query messages from lightningd which we have to filter.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We documented them as deprecated in v23.08 but unintentionally didn't deprecate them in the code. Thus we are starting their actual deprecation cycle from v24.08 release.
Updating removal version for commands `commando-rune`, `commando-listrunes`, and `commando-blacklist`
Changelog-None.
Somehow, this documentation got lost during the Great Rewrite, so
restore that too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-EXPERIMENTAL: JSON-RPC: `offer` removed `@` prefix support from `recurrence_base` (use `recurrence_start_any_period` set to `false`)
We do this by literally creating the modern-style TLV, and pretending we found it in the onion.
This isolates us from messing with any callers, who don't even know.
Co-programmed-with: Alex Myers <alex@endothermic.dev>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Fixes: https://github.com/ElementsProject/lightning/issues/7347
Changelog-Fixed: Protocol: forward legacy non-TLV onions which we removed in 22.11 and spec itself in Feb 2022. Still sent by LND nodes who haven't seen our node_announcement.
This fails, because l2 can't decode the onion:
```
lightningd-2 2024-05-28T21:43:35.137Z DEBUG 0266e4598d1d3c415f572a8488830b60f7e744ed9235eb0b1ba93283b315c03518-chan#1: Rejecting their htlc 0 since onion is unprocessable WIRE_INVALID_ONION_HMAC ss=4202c24ea44d9029a2ea3abb24cded51da93164f8bb5cddce9cc824af9945435
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We used to fire up channeld to send this, but:
1. That's silly, we have all the information to make it ourselves.
2. We didn't do it if there was an error on the channel, which as of 24.02
there always is!
3. When it did work, running channeld *stops* onchaind, indefinitely slowing recovery.
Fixes: https://github.com/Blockstream/greenlight/issues/433
Changelog-Fixed: Protocol: we once again send CHANNEL_REESTABLISH responses on closing channels.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
LDK will pick the *upper* limit (see: https://github.com/lightningdevkit/rust-lightning/issues/3014)!
It should not do this, but since you can set a manual range for mutual close, it's probably better to disable this option for close, as it's even more dangerous than documented.
Changelog-Changed: config/JSON: --ignore-fee-limits / setchannel ignorefeelimits no longer applies to mutual close.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Fixes: https://github.com/ElementsProject/lightning/issues/7242
Thought it's intuitive, it's not documented to, and LDK's current behavior makes it quite dangerous to do so.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This means we can see the values in listconfigs, even if we haven't set
them yet.
In particular, we now see the following:
* autoclean-cycle.value_int=3600
* bitcoin-rpcclienttimeout.value_int=60
* bitcoin-retry-timeout.value_int=60
* funder-max-their-funding.value_str=4294967295sat
* funder-per-channel-min.value_str=10000sat
* funder-reserve-tank.value_str=0sat
* funder-fund-probability.value_int=100
Changelog-Changed: plugins: libplugin now shows plugin option default values (where they're non-trivial)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This demonstrates a number of issues reported by Carla: no surprise
since there was no test!
(We create a one-hop blinded path when we only have private channels).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Our fetchinvoice always creates a reply path which terminates at their peer,
so we need a dev overrride for that.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The feerate security margin is a multiplicative factor applied to the
feerate of some transactions in order to guarantee that the
transaction remains publishable and has a sufficient chance of being
confirmed, that we can base some of our decisions on that.
The multiplicative factor is >=1 and was so far a constant 2. This
might have been sensible in the low-fee environment, where the fees
are expected to oscillate, and almost guaranteeing that we will
eventually have rising feerates but in high-fee environments that is
no longer the case, and the 100% margin that the multiplicator 2
brings is excessive. We therefore opt to start out with 100%, then
linearly interpolate up to a given maxfeerate (which does not have to
be a real feerate ever reached, it just indicates the feerate after
which we apply the constant 10% margin.
Fixes#6974Closes#6976
[Fixed up all the other changes required, including spendable calcualtion
comments and unit test and pytest tests --RR]
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-EXPERIMENTAL: offers: we now understand blinded paths which use a short-channel-id(+direction) as entry point.
We don't actually support it yet, but this threads through the type change,
puts it in "decode" etc.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
And deprecate the --max-locktime-blocks which allows them to set it.
Hilariously, the spec misspells CLTV as CTLV at one point, so we work around it for now.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We expect:
UNUSUAL.*WARNING: we have 1 channels but can file descriptors limited to 65536
We get:
lightningd: WARNING: we have 1 channels but can file descriptors limited to 32768!
This is strange, since the first line is from Python's hard limit. Presumably something is restricting the fd limit of children
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
1024 is a common limit, and people are starting to hit that many channels, so we should increase it: twice the number of channels seems reasonable, though we only do this at restart time.
Changelog-Changed: lightningd: we now try to increase the number of file descriptors, if it's less than twice the number of channels at startup (and log if we cannot!).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Listpeerchannels would update the local channel information setting the
liquidity in the outgoing channel to known_min=known_max=capacity,
when in fact it should be known_min=known_max=spendable.
This didn't trigger the bug, but worth explicitly testing: we spend a
to-remote output from a previous unilateral close, to spend an anchor
on a current unilateral close.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We wire through --dev options into the hsmd, and test preapprove accept and deby
with both old and new protocols.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now we can check that the setconfig would actually parse. We do this
by handing NULL to the calback to indicate it's a test, which may not
work in general but works for now.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: `check` `setconfig` now checks that the new config setting would be valid.
We need to pass through setconfig in check mode, then we need to have libplugin
add (and use!) a `check_only` parameter to all option setting.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: JSON-RPC: `check` `setconfig` on plugin options can now check the config value would be accepted.
Without this, everything came out as level INFO.
Changelog-Fixed: pyln-client: Fix Plugin.notify_message() not to ignore `level` parameter.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This test reproduces a bug in the `cln-plugin`-crate.
Core Lightning supports a wildcard `*` that plugins can use to
subscribe to all notifications.
However, `cln-plugin` does not support this case.
It allows the developer to subscribe `*`.
But the plug-in crashes when the first notification is received
Uncommited channels are missing several fields which would normally be
populated by an active channel. This simply skips them for the purposes
of finding a route. The particular culprit was:
{
"peer_id": "038cd9f3679d5b39bb2105978467918d549572de472f07dd729e37c7a6377d41d5",
"peer_connected": true,
"state": "OPENINGD",
"owner": "lightning_openingd",
"opener": "local",
"to_us_msat": 8317559000,
"total_msat": 8317559000,
"features": [
"option_static_remotekey",
"option_anchors_zero_fee_htlc_tx"
]
}
Fixes#7197 - SEGV in direct_pay_listpeerchannels when field private missing
Changelog-Fixed: Fixed crash in pay plugin caused by parsing uncommitted dual open channels
Changelog-Changed: Plugins: bcli: Add a path that tries to fetch blocks
from a peer if we can not find them locally.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
Rather than `allow_broken_log`, we have `broken_log` which is a regex
indicating what log lines are expected. This tightens our tests
significantly, as it will catch *unexpected* brokenness.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Apparently NixOS didn't have Python (sometimes!) so let's not assume.
By reusing the JSON "parsing" code from cowsay, we can self-disable
to handle this case.
Reported-by: Shahana Farooqui <shahana.farooqui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: Plugins: `clnrest` now correctly self-disables if Python not present at all.
Tests whether we're able to route around a low-fee channel in the graph. We
should be able to find the more expensive route if the cheaper route is
depleted.
In the Min. Cost Flow solver we put a constraint on arcs capacities,
such that the total flow that can be allocated through a channel does
not exceed htlc_max. This solves two previous problem of the htlc_max
constraint at the MCF level.
Add spendable/receivable and is_local fields to the callback function
used in gossmods_from_listpeerchannels. This allows to do more fine
grained use of the listpeerchannels call.
The first use case is renepay, for which we need to ignore the htlc_max
of the local channels only.
I was trying to debug test_zeroconf_open and getting very confused.
The reason: l0 is lightning-1, l1 is lightning-2, etc! And there are
two other tests where an l0 has been added at the front: fix them all
to avoid future confusion!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This allows for faster startup for Greenlight. We still require full sync
before *incoming* htlcs, and onchain operations.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
LDK doesn't set this feature if they don't have any useful gossip (mobile nodes)
and it was agreed at the spec meeting that we should repurpose this feature
to mean "I don't have any useful gossip".
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
These were deprecated in v22.08 (invoice hook) and v0.8 (htlc_accepted hook), and
marked EOL in v23.02.
*PLEASE* complain if this breaks things for you: it's kind of a test canary of
the deprecation system!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: Plugins: `invoice_payment` and `htlc_accepted` hook `failure_code` response (derepcated v22.08 and v0.8, EOL v23.02)
Indeed, we now need to adjust our tests to use --i-promise-to-fix-broken-api-user
for the specific APIs past end-of-life.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We were duplicating the command name (e.g. "autocleaninvoice.autocleaninvoice"), and also not
handling listconfigs if they specified the (currently unused!) i-promise-to-fix-broken-api-user
option, which we are going to use next patch.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We tried to put *everything* into the "all" output, which didn't work
if there were other outputs!
Fixes: https://github.com/ElementsProject/lightning/issues/6664
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: JSON-RPC: `multifundchannel` with `all` as an amount works as expected.
It's a u64, we should pass by copy. This is a big sweeping change,
but mainly mechanical (change one, compile, fix breakage, repeat).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: runes: named parameters (e.g. `pnameamountmsat`) no longer need to remove underscores (i.e. `pnameamount_msat` now works as expected).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Rather than speaking 'rune' we should speak english in error messages.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reported-by: @ShahanaFarooqui
This reveals an inadequacy in our rune error reporting:
we complain a missing parameter is "not an integer field" instead
of "not present":
```
# Rune requires amount_msat < 10,000!
> with pytest.raises(RpcError, match='Not permitted: pnameamountmsat is not present') as exc_info:
E AssertionError: Regex pattern did not match.
E Regex: 'Not permitted: pnameamountmsat is not present'
E Input: "RPC call failed: method: checkrune, payload: {'nodeid': '0266e4598d1d3c415f572a8488830b60f7e744ed9235eb0b1ba93283b315c03518', 'rune': 'b3hXuEM7Pqzk-C7HUw83xzvHOV7fmuGaWjdo-wHdfg89MCZtZXRob2Q9cGF5JnBuYW1lYW1vdW50bXNhdDwxMDAwMA==', 'method': 'pay', 'params': {'bolt11': 'lnbcrt123n1pj7flqdsp5ndqgxpwk2hf50gzm0d4ssgjnd90cwkrc8udh7lfr5x583jms7yqspp5kn5stlnkv70celgw4vmdva9m7a57drd2403vnx4whq2p5nawkh3sdq5v3jhxcmjd9c8g6t0dccsxqyjw5qcqp99qxpqysgqhrgp7wp640gyujxk0mz4l6e6dxmqp7fz8pnnpnnqjfxg2scvuzfpwlxrj332u72p5g709eqr8rwaueruce84h0qmh6kc5c2zxgg9q4qps4cu8k'}}, error: {'code': 1502, 'message': 'Not permitted: pnameamountmsat is not an integer field'}"
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
a.k.a. "Pay with a friend!".
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: JSON-RPC: `pay` has a new parameter `partial_msat` to only pay part of an invoice (someone else presumably will pay the rest at the same time!)
Suggested-by: Calle
- Updated doc/Makefile for generating schemas/lightning-sql.json
- Read the lightning-sql template from schemas/lightning-sql-template.json
- Generate sql tables data and merge it with json object read from above template
- Save final merged object in schemas/lightning-sql.json
- Updated doc/Makefile to auto generate lightning-sql.7.md and lightning-sql.7
- Deleted schemas/lightning-sql.json and adding it into .gitignore
- Added lightning-sql specific titles in the fromschema.py script
- Fixed test_sql by changing the sequence for listpeerchannels `reestablished` field
- Ignoring lightning-sql.json for msggen schema.json because it is auto generated by sql plugin and lightning-sql-template.json.
The amount is set not to crash by default, but run
"common/test/run-route-infloop 8388607" and you'll see a crash.
Sorry about the 7MB blob, but this testing was quite revealing and
I consider it worth adding.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Watchtowers changed the code so that we *always* have a channel->shutdown_scriptpubkey[LOCAL]
(see new_channel()). The previous code had several problems:
1. It tested this for NULL, unnecessarily.
2. It allowed overriding if it was a default, *even* if we were already using it.
3. If the peer opened without option_shutdown_anysegwit, but upgraded before we closed,
we would not recognize the default.
4. It set the final scriptpubkey (and other things!) even if the command failed.
Changelog-Fixed: JSON-RPC: `close` with `destination` works even if prior `destination` was rejected.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Whenever there is a payment failure that requires gossip update, for
example changing the fee rates of remote channels, we call addgossip.
For renepay to consider this changes in the coming payment attempts, it
must update gossmap.
- previous pending sendpays must add up so that the plugin tries to pay
the rest of the amount,
- avoid groupid, partid collisions,
- add shadow fees if the option is set and the payment amount - total
delivering = 0
- add a test,
- also fix a buggy shadow routing test
This avoids us gossiping about nodes which don't have live channels.
Interstingly, we previously tested that we *did* gossip such node
announcements, and now we fix that test.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Add checking for and sending tx_abort to channeld.
When receiving it we first ACK it back, send a request to restart to lightningd, and then shutdown channeld.
We also must update the splice tests that relied on reconnect checks for splice warnings (as some are now tx_aborts instead).
If we try to reuse the same UTXO for the HTLC as the anchor, it will clash.
One block later we can spend the anchor change, all good.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We still want to test non-anchor channels, as we still support them, but
we've made it non-experimental. To test non-anchor channels, we
use dev-force-features: -23.
Changelog-Added: Protocol: `option_anchors_zero_fee_htlc_tx` enabled, no longer experimental.
Changelog-Changed: Config: `experimental-anchors` now does nothing (it's enabled by default).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Header from folded patch 'fixup!_options__make_anchors_enabled_by_default,_ignore_experimental-anchors.patch':
fixup! options: make anchors enabled by default, ignore experimental-anchors.
This uses pip to install to a venv when the dependencies are specified
with requirements.txt and poetry when it's present on the system and the
plugin has a pyproject.toml.
The directory structure will be:
reckless/
source/ (original plugin code here)
plugin_entrypoint (original entrypoint)
plugin_name (symlink or wrapper to activate venv)
.metadata (installation information)
.venv/ (python virtual environment)
The wrapper matches the naming of the original plugin entrypoint. The
shebang is modified to use the virtual environment's python binary,
then the original plugin is imported as a module and executed as though
it was run directly.
Changelog-Changed: reckless installs python plugins in virtual environments
Christian points out that this makes the type harder, and it's just awkward.
Changelog-EXPERIMENTAL: JSON-RPC: Deprecated `offer` parameter `recurrence_base` with `@` prefix: use `recurrence_start_any_period`.
Changelog-EXPERIMENTAL: JSON-RPC: Added `offer` parameter `recurrence_start_any_period`.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
For better performance it is recommended to use the modern OpenSSL
EVP_MD_fetch API to load digest algorithms (i.e. explicit fetching),
instead of the older implicit fetching API.
As a side effect, using this API seems to avoid memory leaks with some
versions of OpenSSL.
We can get bad gossip if a node processes a gossip message after we've closed:
```
_________________________________________ ERROR at teardown of test_closing_specified_destination _________________________________________
...
> raise ValueError(str(errors))
E ValueError:
E Node errors:
E - lightningd-1: had warning messages
E - lightningd-4: had bad gossip messages
E Global errors:
...
lightningd-1 2024-02-03T00:29:02.299Z INFO 0382ce59ebf18be7d84677c2e35f23294b9992ceca95491fcf8a56c6cb2d9de199-connectd: Received WIRE_WARNING: WARNING: channel_announcement: no unspent txout 105x1x0
lightningd-1 2024-02-03T00:29:02.300Z DEBUG 0382ce59ebf18be7d84677c2e35f23294b9992ceca95491fcf8a56c6cb2d9de199-connectd: peer_in WIRE_WARNING
lightningd-1 2024-02-03T00:29:02.300Z INFO 0382ce59ebf18be7d84677c2e35f23294b9992ceca95491fcf8a56c6cb2d9de199-connectd: Received WIRE_WARNING: WARNING: channel_announcement: no unspent txout 103x1x0
lightningd-1 2024-02-03T00:29:02.339Z DEBUG 035d2b1192dfba134e10e540875d366ebc8bc353d5aa766b80c090b39c3a5d885d-connectd: peer_in WIRE_WARNING
lightningd-1 2024-02-03T00:29:02.339Z INFO 035d2b1192dfba134e10e540875d366ebc8bc353d5aa766b80c090b39c3a5d885d-connectd: Received WIRE_WARNING: WARNING: channel_announcement: no unspent txout 103x1x0
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Instead of "new" and "load", we don't really need to "load" anything,
so do everything in gossip_store_new.
Have it do the compaction/rewrite, and collect the dying records
The gossip_store_load is now basically a noop, since gossmap
does that.
gossipd removes a pile of routines dealing with messages,
in favor of just handing them to gossmap_manage.
The stub gossmap_manage constructor is removed entirely.
We simplified behaviour around channel_announcements with
no channel update: we now add them to the store, and go
back to fix the timestamp later. This changes a test,
which explicitly tests for the old behaviour.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We weakened this progressively over time, and gossip v1.5 makes spam
impossible by protocol, so we can wait until then.
Removing this code simplifies things a great deal!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: Protocol: we no longer ratelimit gossip messages by channel, making our code far simpler.
We never enabled it, because we seemed to be eliminating valid
channels. We discard zombie-marked records on loading.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It was an obscure dev command, as it never worked reliably.
It would be much easier to re-implement once this is done.
This turned out to reveal a tiny leak on
tests/test_gossip.py::test_gossip_store_load_amount_truncated where we
didn't immedately free chan_ann if it was dangling.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
update_count is just the count of the records for a tx. To calculate onchain fees
for an account we must sum all credits vs debits. We don't need to GROUP BY update_count
nor ORDER BY update_count since it is just a running index of updates to this tx.
Remove grouping by update_count which resulted in a crash due to bad arithmetic
caused by fee calculation returned rows not being consolidated.
Remove xfail.
```
2024-02-02T10:30:38.6479143Z __________________ ERROR at teardown of test_multifunding_one __________________
...
2024-02-02T10:30:38.6491895Z > raise ValueError(str(errors))
2024-02-02T10:30:38.6492208Z E ValueError:
2024-02-02T10:30:38.6492465Z E Node errors:
2024-02-02T10:30:38.6492833Z E - lightningd-2: had bad gossip messages
2024-02-02T10:30:38.6493535Z E Global errors:
...
2024-02-02T10:30:38.7458545Z lightningd-2 2024-02-02T10:25:48.410Z DEBUG 0266e4598d1d3c415f572a8488830b60f7e744ed9235eb0b1ba93283b315c03518-gossipd: Ignoring future channel_announcment for 103x2x1 (current block 102)
2024-02-02T10:30:38.7460140Z lightningd-2 2024-02-02T10:25:48.411Z UNUSUAL lightningd: Bad gossip order: could not find channel 103x2x1 for peer's channel update
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
