And add contrib/pyln-testing/pyln/testing/grpc2py.py since we didn't previously know
how to build it!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
See the section headed "Rules with Grouped Targets" on the Texinfo page
`(make)Multiple Targets`.
Without this fix, Make does not know that these recipes unconditionally
make *all* of their named targets regardless of which target triggers
their execution, and Make will blissfully execute multiple instances of
any such recipe in parallel, not only wasting CPU cycles but potentially
producing incorrect results if the recipe is not atomic in its effects
on the file system. With this fix, Make understands that it need only
execute such a recipe once to make all of its targets.
In pursuit of the above, move and combine two redundant msggen recipes
into the top-level Makefile, and populate its grouped targets from the
subordinate Makefiles.
Changelog-None
Add a fuzz test for BOLT 8 message encryption and decryption. The fuzz
test is based on the unit test at common/test/run-cryptomsg.c and uses a
static initial state with fuzzer-generated messages to encrypt or
decrypt.
This header will be used by multiple fuzz targets to fuzz Acts 1, 2, and
3 of the BOLT 8 handshake.
We could make this header into a full library, but considering its
narrow use let's try not to over-engineer things.
It seems to reliably be getting a SIGTERM after 17-18 minutes:
```
[gw1] [ 91%] PASSED tests/test_plugin.py::test_forward_event_notification
Error: Process completed with exit code 143.
```
Perhaps this is out of mem? So, try -n2. We also make the configure
line explicit, rather than relying on environment vars (we should probably
do this for the other cases, too)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Since we didn't hash the descriptions properly (see previous commit), we
cannot immediately deprecate omitting the descriptions (since you'd
have to omit them for backwards compat!).
And move the "must have description or hash" test into bolt11.c core.
Changelog-Deprecated: `pay` has *undeprecated* paying a description-hash invoice without providing the description.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This means we need to push off requring this for another full deprecation cycle!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: JSON-RPC: `pay` and `decodepay` with description now correctly handle JSON escapes (e.g " inside description)
This is almost certainly because the HTLCs are not fully settled, so wait for that:
```
2023-06-20T11:37:56.2332158Z assert apys_2[0]['our_start_balance_msat'] == Millisatoshi(0)
2023-06-20T11:37:56.2332443Z > assert apys_1[0]['routed_out_msat'] == apys_2[0]['routed_in_msat']
2023-06-20T11:37:56.2332571Z E assert 1892216msat == 2810170msat
2023-06-20T11:37:56.2332580Z
2023-06-20T11:37:56.2332717Z tests/test_pay.py:81: AssertionError
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If we reconnect before the channel is completely closed, we might get
a "reconnected" message, so mine a block after and make sure it's
processed.
```
2023-06-20T11:37:56.1302058Z if errors.has_errors():
2023-06-20T11:37:56.1302648Z # Format a nice list of everything that went wrong and raise an exception
2023-06-20T11:37:56.1303781Z request.node.has_errors = True
2023-06-20T11:37:56.1304091Z > raise ValueError(str(errors))
2023-06-20T11:37:56.1304370Z E ValueError:
2023-06-20T11:37:56.1304624Z E Node errors:
2023-06-20T11:37:56.1305042Z E - lightningd-2: had unexpected reconnections
2023-06-20T11:37:56.1305340Z E Global errors:
```
...
```
2023-06-20T11:37:56.1960525Z lightningd-2 2023-06-20T11:21:28.638Z DEBUG 0266e4598d1d3c415f572a8488830b60f7e744ed9235eb0b1ba93283b315c03518-chan#2: Peer has reconnected, state CLOSINGD_SIGEXCHANGE: connecting subd
```
I added a plugin arg and was surprised that compile didn't break.
This is because typesafe_cb et al are conditional casts: if the type
isn't as expected it has no effect, but we're passing plugin_option() through
varargs, so everything is accepted!
Add a noop inline to check type, and fix up the two cases where we
used `const char *` instead of `char *`.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If it's a plugin opt, we'll need a callback, so reshuffle logic. Also
add infra to map option name to plugin and option.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Do it slightly intelligently, so if we had set previously using setconfig
we don't keep appending new ones, but replace it in-place.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Currently only implemented for min-capacity-sat.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: JSON-RPC: new command `setconfig` allows a limited number of configuration settings to be changed without restart.
Previously, if these failed we always exited; once we have dymamic
configs this would be a (tiny) memory leak, so use tmpctx.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
To test, we do min-capacity-sat which is simple. We also update the
listconfigs man page which contained some obsolete information.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We get "disconnected during connection" if we haven't finished processing
the connection when the peer sends error and hangs up.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We re-enable sendrawtransaction then mine a block to kick off RBF, but there's
a race where it can get a tx in that block, and then we timeout waiting for
both txs to get into the next block.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This isn't the cause of the test_gossip_ratelimit flake I saw (since
the final gossip msg clearly was received), but it's still good to fix
since it means we might not send the final messages.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We were seeing hangs in disconnect in
tests/test_connection.py::test_feerate_stress, and looking at the logs
it's because we're not actually telling connectd to disconnect.
These days, we have a connect counter, so connectd knows to ignore it
if we simply haven't read the message about it already disconnecting.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
build.sh did this, but we didn't carry it across to the new test
matrix, leading to spurious CI failures:
```
lightningd-1 2023-06-06T16:15:34.931Z DEBUG plugin-bookkeeper: Setting up database at ***localhost:39061/accounts_0_7ccmg745
lightningd-1 2023-06-06T16:15:34.931Z INFO plugin-bookkeeper: Creating database
lightningd-1 2023-06-06T16:15:34.931Z **BROKEN** plugin-bookkeeper: Error vacuuming db: VACUUM command failed: ERROR: deadlock detected\nDETAIL: Process 77248 waits for AccessShareLock on relation 1260 of database 0; blocked by process 77414.\nProcess 77414 waits for RowExclusiveLock on relation 1214 of database 0; blocked by process 77248.\nHINT: See server log for query details.\n
lightningd-1 2023-06-06T16:15:34.931Z INFO plugin-bookkeeper: Killing plugin: exited before replying to init
lightningd-1 2023-06-06T16:15:34.931Z **BROKEN** plugin-bookkeeper: Plugin marked as important, shutting down lightningd!
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
