Instead of making it ourselves, lightningd does it. Now we only have
two cases of failed htlcs: completely malformed (BADONION), and with
an already-wrapped onion reply to send.
This makes channeld's job much simpler.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I hadn't realized that lightningd asks gossipd every time we forward
a payment. But I'm going to abuse it here to get the latest channel_update,
otherwise (as lightningd takes over error message generation) lightningd
needs to do an async request at various painful points.
So have gossipd tell us the lastest update (stripped so compatible with
the strange in-onion-error format).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Turn it into temporary node failure: this only happens if we restart
with a failed htlc in, but it's clearer and more robust to handle it
generically.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
For incoming htlcs, we need failure details in case we need to
re-xmit them. But for outgoing htlcs, lightningd is telling us it
already knows they've failed, so we just need to flag them failed
and don't need the details.
Internally, we set the ->fail to a dummy non-NULL value; this is
cleaned up next.
This matters for the next patch, which moves onion handling into
lightningd.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
1. forward_htlc sets hout to NULL.
2. forward_htlc passes &hout to send_htlc_out.
3. forward_htlc checks the failcode and frees(NULL) and sets hout to NULL
(again). This in fact covers every failcode which send_htlc_out returns.
We should ensure send_htlc_out sets *houtp to NULL on failure; in fact,
both callers pass houtp, so we can make it unconditional.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We don't compile with NDEBUG defined, but if we did, this code would
vanish. I did a quick audit, inspired by @ZmnSCPxj.
I actually hacked up something to compile with NDEBUG (many unused vars
resulted, and of course unit tests are allowed to rely on assert()), and
after this the testsuite still passes.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I reproduced this by putting a sleep(60) in the pay plugin, then
'lightning-cli pay', 'lightning-cli plugin stop pay' and then ^C
the `lightning-cli pay`:
2020-02-14T00:33:11.217Z INFO plugin-pay: Killing plugin: pay stopped by lightningd via RPC
2020-02-14T00:33:15.250Z DEBUG lightningd: Still waiting for initial block download
==5157== Invalid read of size 8
==5157== at 0x12A29C: destroy_jcon (jsonrpc.c:149)
==5157== by 0x1C6F2A: notify (tal.c:235)
==5157== by 0x1C7441: del_tree (tal.c:397)
==5157== by 0x1C7493: del_tree (tal.c:407)
==5157== by 0x1C77DD: tal_free (tal.c:481)
==5157== by 0x1B7380: io_close (io.c:450)
==5157== by 0x1B71A7: do_plan (io.c:401)
==5157== by 0x1B7214: io_ready (io.c:417)
==5157== by 0x1B94AC: io_loop (poll.c:445)
==5157== by 0x1291C9: io_loop_with_timers (io_loop_with_timers.c:24)
==5157== by 0x12EC7E: main (lightningd.c:928)
==5157== Address 0x4ebab98 is 40 bytes inside a block of size 88 free'd
==5157== at 0x483BA3F: free (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==5157== by 0x1C750F: del_tree (tal.c:416)
==5157== by 0x1C7493: del_tree (tal.c:407)
==5157== by 0x1C77DD: tal_free (tal.c:481)
==5157== by 0x153856: clear_plugin (plugin_control.c:209)
==5157== by 0x1538FF: plugin_dynamic_stop (plugin_control.c:225)
==5157== by 0x153C51: json_plugin_control (plugin_control.c:295)
==5157== by 0x12B4EC: command_exec (jsonrpc.c:588)
==5157== by 0x12B8AB: rpc_command_hook_callback (jsonrpc.c:679)
==5157== by 0x154575: plugin_hook_call_ (plugin_hook.c:170)
==5157== by 0x12BCD3: plugin_hook_call_rpc_command (jsonrpc.c:756)
==5157== by 0x12BD04: call_rpc_command_hook (jsonrpc.c:764)
==5157== Block was alloc'd at
==5157== at 0x483A7F3: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==5157== by 0x1C6F98: allocate (tal.c:245)
==5157== by 0x1C7559: tal_alloc_ (tal.c:423)
==5157== by 0x15135A: plugin_rpcmethod_add (plugin.c:706)
==5157== by 0x151600: plugin_rpcmethods_add (plugin.c:756)
==5157== by 0x151BDD: plugin_parse_getmanifest_response (plugin.c:893)
==5157== by 0x151C9C: plugin_manifest_cb (plugin.c:915)
==5157== by 0x14FFB9: plugin_response_handle (plugin.c:258)
==5157== by 0x150165: plugin_read_json_one (plugin.c:356)
==5157== by 0x1502BC: plugin_read_json (plugin.c:388)
==5157== by 0x1B65ED: next_plan (io.c:59)
==5157== by 0x1B71D2: do_plan (io.c:407)
Fixes: #3509
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If the peer is not connected, or other error which means we don't
actually create an outgoing HTLC, we don't record the
short_channel_id. This is unhelpful!
Pass the scid down to the wallet code, and explicitly hand the
scid and amount down to the notification code rather than handing it
the htlc_out (which it doesn't need).
Changelog-Changed: JSON API: `listforwards` now shows `out_channel` even if we couldn't forward.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This avoids the getblockhash+getblock, and more importantly that was the
last functionality making use of bitcoind_getrawblock() and bitcoin_getblockhash(),
so we can also get rid of them.
This adds `getchaininfo` and `getrawblockbyheight` handling lightningd-side,
and use them in setup_topology().
We then remove legacy bitcoind_getblockcount() (we already get the count in
`getchaininfo`), bitcoind_getblockchaininfo() (it was only used in setup_topology()),
and wait_for_bitcoind() (this was specific to bitcoin-core and we assume our Bitcoin
backend to be functional if the plugin responds to `init`).
Add new check if we're funder trying to add HTLC, keeping us
with enough extra funds to pay for another HTLC the peer might add.
We also need to adjust the spendable_msat calculation, and update
various tests which try to unbalance channels. We eliminate
the now-redundant test_channel_drainage entirely.
Changelog-Fixed: Corner case where channel could become unusable (https://github.com/lightningnetwork/lightning-rfc/issues/728)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Make the `htlc_accepted` hook the first chained hook in our repertoire. The
plugins are called one after the other in order until we have no more plugins
or the HTLC was handled by one of the plugins. If no plugins handles the HTLC
we continue to handle it internally like always.
Handling in this case means the plugin returns either `{"result": "resolve",
...}` or `{"result": "fail", ...}`.
Changelog-Changed: plugin: Multiple plugins can now register for the htlc_accepted hook.
This used to be necessary because we allocated the `plugin_hook_request` off
of the plugin instance (only tal allocated object we could grab at that
time. Now the plugin was replaced by a list, which itself is tal-allocated,
making that workaround pointless, or even wrong once we have multiple plugins
registering for that hook.
We will be using `plugin_hook_call_next` as part of the loop to traverse all
plugins that registered the hook, so group initialization in the init function
and move per-plugin logic into `plugin_hook_call_next`
We are about to call multiple plugins, and we'll have to pass the payload into
each call. Sadly the serialized stream gets consumed during the call, so keep
the unserialized payload around.
Switch from having a single plugin to a list of plugins. If the hook is of
type single we will enforce that constraint on the number of registered
plugins when attempting to add.
The newly introduced type is used to determine what the call semantics of the
hook are. We have `single` corresponding to the old behavior, as well as
`chain` which allows multiple plugins to register for the hook, and they are
then called sequentially (if all plugins return `{"result": "continue"}`) or
exit the chain if the hook event was handled.
This is the last venue we need to add custom featurebits to, so we also unmark
the test as xfail.
Changelog-Added: plugin: Plugins can now signal support for experimental protocol extensions by registering featurebits for `node_announcement`s, the connection handshake, and for invoices. For now this is limited to non-dynamic plugins only
The `init_featurebits` are computed at startup, and then cached
indefinitely. They are then used whenever a new `init` handshake is performed.
We could add a new message to push updates to `connectd` whenever a plugin is
added or removed, but that's up for discussion.
Break out a method for canceling a channel that will either
loop through contacting the peer to tell them of the error or
just directly cleans up if the peer is currently disconnected.
restrict fundchannel_cancel usage to only the opener side
Changelog-Changed: Only the opener of a fundchannel can cancel the channel open with fundchannel_cancel
As a separated commit because it was pre-existent (changelog + xfail test).
This also fix a logical problem in lightningd/plugin_control: we were
assuming a plugin started with 'plugin start' but which did not comport
a 'dynamic' entry in its manifest to be dynamic, though it should have
been treated as static.
Changelog-fixed: plugins: Dynamic C plugins can now be managed when lightningd is up
Now that we have json_stream in common/, we can move all the related
helpers from lightningd/json to common/json. This way everyone can
benefit of them (including libplugin, the plugins themselves,
potentially lightning-cli), not lightningd alone!
Note that the Makefile of the common/test/ had to be modified, because
the new helpers make use of common/wireaddr... Which turns out to
\#include <lightingd/lightningd.h> ! So we couldnt just include the .c
and add mocks if we redefined some structs (hello run-param).
Changelog-Changed: The hooks `db_write`, `invoice_payment`, and `rpc_command` now accept `{ "result": "continue" }` to mean "do default action", in addition to `true` (`db_write`), `{}` (`invoice_payment`), and `{"continue": true}` (`rpc_command`). The older "default" indicators are now deprecated and are now recognized only if `--deprecated-apis` is set.
Changelog-Added: lightningd: Added --subdaemon command to allow alternate subdaemons.
[ Wow, that was mammoth; 44 comments over 12 commits. Feels almost unfair to squash it into one commit, so I wanted to note @ksedgwic's perseverence here! --RR ]
Using it with a different value to the amount sent causes a crash in 0.8.0,
which is effectively deprecating it, so let's disallow it now.
Changelog-Changed: If the optional `msatoshi` param to sendpay for non-MPP is set, it must be the exact amount sent to the final recipient.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We tag them with specific versions when they're experimental,
but do a poor job of cleaning them up (and thus ensuring they're
checked!) afterwards.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
