The general pattern for xxx_new is that it should populate all
fields, for encapsulation and so you never can have a half-formed
object.
This means a fair bit of work for now, but it pays off in the next
patch.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
You cannot refresh the gossmap with localmods applied, nor apply localmods
when others have applied localmods in the same process.
There are optimizations we could do, but for now always apply/unapply before
querying gossmap.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We've stomped errno, so if exec fails we don't get a reliable result:
```
2023-08-07T17:58:45.713Z **BROKEN** plugin-bcli: bitcoin-cli exec failed: Bad file descriptor
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
See: https://github.com/bitcoindevkit/bdk/issues/1047#issuecomment-1660645669
In general, futures produced by most libraries in the ecosystem of Rust, and bounds placed
on users of famous runtimes like tokio and its spawn method all lack Sync requirements.
Because of this, anyone who creates a callback using any sort of library that returns a
non-Sync future (which most libraries fit this description) inside of it will get some
cryptic error messages (async error messages still leave a lot to be desired).
Removing these Sync requirements will make the library more useful.
Fixes 32-bit builds:
```
In file included from plugins/renepay/pay.c:5:
./plugins/renepay/pay_flow.h: In function 'fmt_payflow_key':
./plugins/renepay/pay_flow.h:54:17: error: format '%ld' expects argument of type 'long int', but argument 4 has type 'u64' {aka 'long long unsigned int'} [-Werror=format=]
54 | "key: groupid=%ld, partid=%ld, payment_hash=%s",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
55 | k->groupid,k->partid,
| ~~~~~~~~~~
| |
| u64 {aka long long unsigned int}
```
etc
It's expressed in bits, but really it's clearer as a quantity, given
how it's used.
Suggested-by: @Lagrang3
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Update the lightningd <-> channeld interface with lots of new commands to needed to facilitate spicing.
Implement the channeld splicing protocol leveraging the interactivetx protocol.
Implement lightningd’s channel_control to support channeld in its splicing efforts.
Changelog-Added: Added the features to enable splicing & resizing of active channels.
Update gossip routiens and various other hecks on the channel state to consider AWAITING_SPLICE to be routable and treated similar to CHANNELD_NORMAL.
Small updates to psbt interface
Changelog-None
- adopt "const <type> *"convention
- remove use_shadow option for some pyln tests
- show prob. information of flows into paynotes
- show prob. of success of entire payment flow in paynotes
- minflow: We were not releasing the memory of flow arrays when replacing
them with a new canditate.
- use memleak_scan_obj in memleak_check
- replace u64 with size_t
Signed-off-by: Lagrang3 <eduardo.quintana@pm.me>
The global is an *internal* hack because dijkstra_item_mover doesn't
take a context arg! It should be used with care.
Easy, since all the accessors exist: we just hand in the struct dijkstra.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
- remove internal gheap checks
- add check for arc_t.chanidx overflow
- remove outdated comments
- check the delta flow bounds before augmenting along a path
- get_flow_paths uses a dynamic tal array instead of a list.
- fix a unit test that depended on the order of returned flows
- fix bug: lightnind doesn't like if I reuse the partid of a failed
flow, therefore use a higher partid than any of the previous attempts.
- plugin_err instead of LOG_BROKEN if sendpay fails and we cannot get a
an error code.
- fix wrong comments.
- remove the background timer.
- This is a bugfix. Previous to this the MCF network was built using the
knowledge of the min and max liquidity but it didn't take into account
pending HTLCs.
- Also remove the min_prob_success option but hardcode a 90% value.
Removing some options that are not relevant to the user, they're kept
for developer mode only:
- base_fee_penalty
- min_prob_success
- prob_cost_factor
- remove heap.h, not used
Signed-off-by: Lagrang3 <eduardo.quintana@pm.me>
This way unreserving the PSBT will work as intended, and we don't have
to keep track of how many times we've called reserved for any one input.
Technically we're supposed to not reserve inputs at *all* while doing
opens, this moves us slightly closer to that.
The alias may not be set for non-alias channels after they
confirm. The other branch is safe because we only consider active
channels.
Changelog-None
Fixes#6450
Caught by leak detection, we just re-assigned this when we retried: sure,
it's temporary, but it's technically a leak.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I shut down bitcoind during a test, and bcli leak reports flooded in.
They're all temporary, but this fixes them.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is the simplest solution, not the best, but there's significant risk in try to remove the "we have a path" assumption in the code pay code.
Includes removing a `tal_steal` which was incorrect: the buffer has the same lifetime as the plugin, so if we steal it then things get messy when we free the struct payment.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: Plugins: `pay` will now pay your own invoices if you try.
This means (temporarily) that blacklisting won't work (fix later), and
means that old-style (commando.py) master-secret-override doesn't work.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: Plugins: `commando` no longer allows datastore ['commando', 'secret'] to override master secret (re-issue runes if you were using that!).
In preparation for going async:
1. Split try_command's tail into a new function called execute_command() after
the rune checks have succeeded.
2. Put all the info execute_command() needs into struct cond_info, to make it
a simple callback style.
So we create new_cond_info() which dynamically allocates `struct cond_info`
and sets the destructor.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We would create a `struct commando` to marshal our incoming messages,
then try_command would create a *new* one. We can simply reuse, but
when I did I noticed a trick: the new one was not in the `incomings`
array, so didn't work towards the ratelimit. So we need to remove it
from `incomings` in `try_command`, but at least it's now explicit.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We used to activate on the first rune creation, but we're no longer in charge
of runes, so we can't make that call.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Debugging a number of payments showed that we sometimes waste a number
of attempts routing through a channel via its alias, rather than its
scid. This is because while we annotate the scid when it has been set,
we do not do so for the alias. The alias then is picked for routing
despite not having enough capacity, failing the attempt locally.
It can also happen that we alternate between scid and alias, doubling
the number of failed attempts before we can make progress.
This patch sets the hint for the alias to a capacity of 0 and disables
it as if the peer were offline. This means when available we'll always
use the scid, which is also far easier to read in the logs.
Changelog-Fixed: pay: We now track spendable amounts when routing on both the local alias as well as the short channel ID
The presplitter modifier would split a payment before trying the first
attempt based on some common sizes. Its goal was to have smaller parts
in flight over different paths, in order to make it more difficult for
a forwarding node to learn payment amount. However it was causing some
issues for direct payments, and estimates on spendable amounts which
considers only the first HTLC being added, but presplitter would
always cause multiple HTLCs to be kicked off, causing the estimate to
be off.
Removing the presplitter fixes this, making draining channels easier,
and worse success rates, due to more HTLCs in flight directly
impacting the changes of getting stuck.
Changelog-Removed: pay: `pay` no longer splits based on common size, as it was causing issues in various scenarios.
