At the moment, we store e.g. WIRE_TEMPORARY_CHANNEL_FAILURE, and then
lightningd has a large demux function which turns that into the correct
error message.
Such an enum demuxer is an anti-pattern.
Instead, store the message directly for output HTLCs; channeld now
sends us an error message rather than an error code.
For input HTLCs we will still need the failure code if the onion was
bad (since we need to prompt channeld to send a completely different
message than normal), though we can (and will!) eliminate its use in
non-BADONION failure cases.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We're going to change our internal structure next, so this is preparation.
We populate existing errors with temporary node failures, for simplicity.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
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>
Added in d901304120, this column is null in old dbs like mine:
2020-02-15T00:08:41.444Z **BROKEN** database: Accessing a null column 12 in query SELECT id, channel_htlc_id, msatoshi, cltv_expiry, hstate, payment_hash, payment_key, routing_onion, failuremsg, malformed_onion, origin_htlc, shared_secret, received_time FROM channel_htlcs WHERE direction= ? AND channel_id= ? AND hstate != ?
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 sets the nLockTime to the tip (and accordingly each input's nSequence to
0xfffffffe) for withdrawal transactions.
Even if the anti fee-sniping argument might not be valid until some time yet,
this makes our regular wallet transactions far less distinguishable from
bitcoind's ones since it now defaults to using native Segwit transactions
(like us). Moreover other wallets are likely to implement this (if they
haven't already).
Changelog-Added: wallet: withdrawal transactions now sets nlocktime to the current tip.
Before this patch we used `int` for error codes. The problem with
`int` is that we try to pass it to/from wire and the size of `int` is
not defined by the standard. So a sender with 4-byte `int` would write
4 bytes to the wire and a receiver with 2-byte `int` (for example) would
read just 2 bytes from the wire.
To resolve this:
* Introduce an error code type with a known size:
`typedef s32 errcode_t`.
* Change all error code macros to constants of type `errcode_t`.
Constants also play better with gdb - it would visualize the name of
the constant instead of the numeric value.
* Change all functions that take error codes to take the new type
`errcode_t` instead of `int`.
* Introduce towire / fromwire functions to send / receive the newly added
type `errcode_t` and use it instead of `towire_int()`.
In addition:
* Remove the now unneeded `towire_int()`.
* Replace a hardcoded error code `-2` with a new constant
`INVOICE_EXPIRED_DURING_WAIT` (903).
Changelog-Changed: The waitinvoice command would now return error code 903 to designate that the invoice expired during wait, instead of the previous -2
This command injects a custom message into the encrypted transport stream to
the peer, allowing users to build custom protocols on top of c-lightning
without requiring any changes to c-lightning itself.
Generally I prefer structures over u8, since the size is enforced at
runtime; and in several places we were doing conversions as the code
using Sphinx does treat struct secret as type of the secret.
Note that passing an array is the same as passing the address, so
changing from 'u8 secret[32]' to 'struct secret secret' means various
'secret' parameters change to '&secret'. Technically, '&secret' also
would have worked before, since '&' is a noop on array, but that's
always seemed a bit weird.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This makes it clear we're dealing with a message which is a wrapped error
reply (needing unwrap_onionreply), not an already-wrapped one.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I really want a type which means "I am a wrapped onion reply" as separate
from "I am a normal wire msg". Currently both user u8 *, and I got very
confused trying to figure out where each one was an unwrapped error msg,
or where it still needed (un)wrapping.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
`wallet_payment_store` would free the `wallet_payment` instance which would
then cause us to reload it from the DB. Instead of doing the store->free->load
dance we now tell `wallet_payment_store` whether it should take ownership and
leave it alone if not.
Passing the payment around instead of referencing it through payment_hash and
partid is a nice side-effect.
The optimistic lock prevents multiple instances of c-lightning making
concurrent modifications to the database. That would be unsafe as it messes up
the state in the DB. The optimistic lock is implemented by checking whether a
gated update on the previous value of the `data_version` actually results in
an update. If that's not the case the DB has been changed under our feet.
The lock provides linearizability of DB modifications: if a database is
changed under the feet of a running process that process will `abort()`, which
from a global point of view is as if it had crashed right after the last
successful commit. Any process that also changed the DB must've started
between the last successful commit and the unsuccessful one since otherwise
its counters would not have matched (which would also have aborted that
transaction). So this reduces all the possible timelines to an equivalent
where the first process died, and the second process recovered from the DB.
This is not that interesting for `sqlite3` where we are also protected via the
PID file, but when running on multiple hosts against the same DB, e.g., with
`postgres`, this protection becomes important.
Changelog-Added: DB: Optimistic logging prevents instances from running concurrently against the same database, providing linear consistency to changes.
This increments the `data_version` upon committing dirty transactions, reads
the last data_version upon startup, and tracks the number in memory in
parallel to the DB (see next commit for rationale).
Changelog-Changed: JSON-RPC: Added a `data_version` field to the `db_write` hook which returns a numeric transaction counter.
This is the final step: we pass the complete fee_states to and from
channeld.
Changelog-Fixed: "Bad commitment signature" closing channels when we sent back-to-back update_fee messages across multiple reconnects.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The upgrade here is a bit tricky: we map the two values into the
feerate_state. This is trivial if they're both the same, but if
they're different we don't know exactly what state they're in (this
being the source of the bug!).
So, we assume that the have received the update and not acked it,
as that would be the normal case.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The invoice_try_pay code now takes a set, rather than a single htlc, but
it's basically the same thing.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is a transient field, so rework things so we don't leave it in
struct htlc_out. Instead, load htlc_in first and connect htlc_out to
them as we go.
This also changes one place where we use it instead of the am_origin
flag.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is in preparation for partial payments. For existing payments,
partid is 0 (to match the corresponding payment).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is in preparation for partial payments. For existing payments,
partid is 0 (arbitrarity) and total_msat is msatoshi.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now "raw_payload" is always the complete string (including realm or length
bytes at the front).
This has several effects:
1. We can receive an decrypt an onion which is grossly malformed.
2. We can still hand this to the htlc_accepted hook.
3. We then fail it unless the htlc_accepted accepts it manually.
4. The createonion API now takes the raw payload, and does not know
anything about "style".
The only caveat is that the sphinx code needs to know the payload
length: we have a call for that, which simply tells it to copy the
entire onion (and treat us as the final node) if it's invalid.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If we can't decode the onion, because the onion got corrupted or we used
`sendonion` without specifying the `shared_secrets` used, the best we can do
is tell the caller instead.
This means that c-lightning can now internally decrypt an eventual error
message, and not force the caller to implement the decryption. The main
difficulty was that we now have a new state (channels and nodes not specified,
while shared_secrets are specified) which needed to be handled.
We are breaking with a couple of assumptions, namely that we have the
`path_secrets` to decode the error onion. If this happens we just want it to
error out.
We don't set the secret to compulsory (yet!) but put code in for the
future. Meanwhile, if there is a secret, check it is correct.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
In a future version, we will use features to insist that payers
provide the secret. In transition, we may have old invoices which
didn't insist on that, so we need to know this on a per-invoice basis.
Not sure if I got the right syntax for adding an empty blob though!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We have consolidated the two functions into a single `route_step_decode`
function, and made it static since we call it in the `process_onionpacket`
function. We remove the two exposed functions since they're no longer useful.
Since the parser itself just parses and doesn't include validation anymore we
need to put that functionality somewhere. The validation consists of enforcing
that the types are in monotonically increasing order without duplicates and
that for the even types we know how to handle it.
This allows finegrained logging control of particular subdaemons or
subsystems.
To do this, we defer setting the logging levels for each log object
until after early argument parsing (since e.g. "bitcoind" log object
is created early).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-changed: Options: log-level can now specify different levels for different subsystems.
1. Printed form is always "[<nodeid>-]<prefix>: <string>"
2. "jcon fd %i" becomes "jsonrpc #%i".
3. "jsonrpc" log is only used once, and is removed.
4. "database" log prefix is use for db accesses.
5. "lightningd(%i)" becomes simply "lightningd" without the pid.
6. The "lightningd_" prefix is stripped from subd log prefixes, and pid removed.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-changed: Logging: formatting made uniform: [NODEID-]SUBSYSTEM: MESSAGE
Changelog-removed: `lightning_` prefixes removed from subdaemon names, including in listpeers `owner` field.
We had a separate logbook for each peer, and copy log entries above
the printable log level into the master logbook. This didn't always
work well, since we didn't dump it on crash for example.
Keep a single global logbook instead, and remove this infrastructure.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
A log can have a default node_id, which can be overridden on a per-entry
basis. This changes the format of logging, so some tests need rework.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The case where this is needed is when the wallet had a forwarded payment
somewhere between commits 66a47d2 (which started tracking forwardings) and
d901304 (which added the `received_time` column). This just emulates the
behavior of sqlite3 for postgres as well.
Signed-off-by: Christian Decker <@cdecker>
Checking on whether we access a null field is ok, but should we crash right
away? Probably not. This reduces the access to a warning on sqlite3 and let's
it continue. We can look for occurences and fix them as they come up and then
re-arm the asserts once we addressed all cases.
We were implicitly relying on sqlite3 behavior that returns the zero-value for
nulled fields when accessing them. This adds the same behavior explicitly to
the DB abstraction in order to reduce `db_column_is_null` checks in the logic,
but still make it evident what is happening here.
Fixes https://github.com/fiatjaf/mcldsp/issues/1
Signed-off-by: Christian Decker <@cdecker>
I had a report of a 0.7.2 user whose node hadn't appeared on 1ml. Their
node_announcement wasn't visible to my node, either.
I suspect this is a consequence of recent version reducing the amount of
gossip they send, as well as large nodes increasingly turning off gossip
altogether from some peers (as we do). We should ignore timestamp filters
for our own channels: the easiest way to do this is to push them out
directly from gossipd (other messages are sent via the store).
We change channeld to wrap the local channel_announcements: previously
we just handed it to gossipd as for any other gossip message we received
from our peer. Now gossipd knows to push it out, as it's local.
This interferes with the logic in tests/test_misc.py::test_htlc_send_timeout
which expects the node_announcement message last, so we generalize
that too.
[ Thanks to @trueptolmy for bugfix! ]
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is mainly an internal-only change, especially since we don't
offer any globalfeatures.
However, LND (as of next release) will offer global features, and also
expect option_static_remotekey to be a *global* feature. So we send
our (merged) feature bitset as both global and local in init, and fold
those bitsets together when we get an init msg.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This patch adds a channel_id parameter to allow for specifying
channels that are lacking a short_channel_id.
Useful in the case where a peer has 1) multiple channels (ONCHAIN etc)
and 2) a channel where the funding transaction hasn't been
broadcast/mined.
Command format: close id [unilateraltimeout] [destination]
Close the channel with peer {id}, forcing a unilateral
close after {unilateraltimeout} seconds if non-zero, and
the to-local output will be sent to {destination}. If
{destination} isn't specified, the default is the address
of lightningd.
Also change the pylightning:
update the `close` API to support `destination` parameter
`shutdown_scriptpubkey[REMOTE]` is original remote_shutdown_scriptpubkey;
`shutdown_scriptpubkey[LOCAL]` is the script used for "to-local" output when `close`. Add the default is generated form `final_key_idx`;
Store `shutdown_scriptpubkey[LOCAL]` into wallet;
Our policy is generally to omit fields which aren't sensible.
Also, @niftynei points out the spacing in for loops.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This triple join should be efficient to read, and to process. We have a
one-to-many (tx-to-annotations), followed by a
one-to-one (annotation-to-channel) join, so we are limited to annotations x
transactions results.
We have split the iteration over the txs and the output in different
functions, so pushing the annotation down, while keeping the transaction
addition atop. This showcases the need to not have the txid reference the
transactions.id in the DB: we annotate in a function that doesn't have the tx
index context, but only add the TX after we have finished extracting.
Currently the only source for amount_asset is the value getter on a tx output,
and we don't hand it too far around (mainly ignoring it if it isn't the
chain's main currency). Eventually we could bubble them up to the wallet, use
them to select outputs or actually support assets in the channels.
Since we don't hand them around too widely I thought it was ok for them to be
pass-by-value rather than having to allocate them and pass them around by
reference. They're just 41 bytes currently so the overhead should be ok.
Signed-off-by: Christian Decker <@cdecker>
In elements we add an explicit fee output, if we don't consider it when
selecting coins, we end up underpaying the fees.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
Skipping coinbase transactions and ensuring that the transaction is serialized
correctly when sending it onwards.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
With enable-autotor-v2 defined in cmdline the default behavior to create
v3 onions with the tor service call, is set to v2 onions.
Signed-off-by: Saibato <saibato.naga@pm.me>
The DB field type has to match the size of the accessor-type, and we had to
split the `REPLACE INTO` and `INSERT INTO OR IGNORE` queries into two
queries (update and insert if not updated) since there is no portable UPSERT
operation, but impact should be minimal.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
sqlite3 was forgiving, postgres isn't, so let's make sure we use the strictest
field type possible, relaxing when rewriting.
The commit consists just of the following mapping
- INTEGER -> BIGSERIAL if it is the primary key
- INTEGER -> BIGINT if it is an amount or a reference to a primary key
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This was already done in `db_step` but `db_count_changes` and
`db_last_insert_id` also rely on the statement being executed. Furthermore we
now check that the statement was executed before freeing it, so it can't
happen that we dispose of a statement we meant to execute but forgot.
The combination of these could be used to replace the pending_statement
tracking based on lists, since we now make sure to execute all statements and
we use the memleak checker to make sure we don't keep a statement in memory.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
sqlite3 will just report 0 for anything that it thinks should be numeric, or
is accessed using a numeric accessor. Postgres does not, so we need to check
for is_null before trying to read it.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
sqlite3 doesn't really do any validation whatsoever, and there is no
difference between 64bit and 32bit numbers. Posgtres on the other hand gets
very upset if the size doesn't match.
This commit swaps out handwavy types with the ones that should be there :-)
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This was weird right from the start, so we just split the table into integers
and blobs, so each column has a well-defined format. It is also required for
postgres not to cry about explicit casts in the `paramTypes` array.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
The first ever query to check if the version DB exists may fail. We allow
this, but we need to restart the DB transaction since postgres fails the
current transaction and rolls back any changes.
This just commits (and fails) and starts a new transaction so the rest of the
migration can continue.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
Needed to change a couple of migrations. The changes are mostly innocuous:
- changing BLOB to TEXT for short_channel_ids which is the correct type
anyway, and sqlite3 treats them the same anyway.
- Use `int` for version since the byte representation is checked by postgres.
- Change anything that is INT, but will be bound to u64 to BIGINT (again
postgres checks these more carefully than sqlite3).
Two migrations were replaced with dummy values, since they are buried deep
enough, and I found no portable way of expressing `strftime()` and `INSERT OR
IGNORE`.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
Using a generated identifier with filename and line proved to be brittle since
compilers assign the __LINE__ macro differently on multi-line macro
invocations.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This is dangerous but needed since postgres is not as forgiving about
unsatisfied foreign key constraints even while in a DB transaction.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
We used to do some of the setup work in db.c, which is now free of any
sqlite3-specific code. In addition we also switch over to fully qualified DSNs
to specify the location of the wallet.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This avoids having to correlate with listpeers for the most pertinent
information.
This API predates plugins, otherwise we'd have listutxos and listpeers
and this would simply combine them appropriately. Still, it exists so
there's little reason not to make it more friendly.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Allow a user to select the utxo set that will be added to a
transaction, via the `utxos` parameter. Optional.
Format for utxos should be of the form ["txid:vout","..."]
Rather than reaching into data structures, let them register their own
callbacks. This avoids us having to expose "memleak_remove_xxx"
functions, and call them manually.
Under the hood, this is done by having a specially-named tal child of
the thing we want to assist, containing the callback.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We're about to remove them.
Includes fix to sqlite3_bind_short_channel_id to not assume `id` is a
tal object.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now that all the users are migrated to the abstraction layer we can remove the
legacy implementation.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
We are about to delete all the `sqlite3`-specific code from `db.c` and this is
one of the last uses of the old interface.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
It's better to let the driver decide when and how to expand. It can then
report the expanded statement back to the dispatch through the
`db_changes_add` function.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
We now have a much stronger consistency check from the combination of
transaction wrapping, tal memory leak detection. Tramsaction wrapping ensures
that each statement is executed before the transaction is committed. The
commit is also driven by the `io_loop`, which means that it is no longer
possible for us to have statements outside of transactions and transactions
are guaranteed to commit at the round's end.
By adding the tal-awareness we can also get a much better indication as to
whether we have un-freed statements flying around, which we can test at the
end of the round as well.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This is likely the last part we need to completely encapsulate the part of the
sqlite3 API that we were using. Like the `db_count_changes` call I decided to
pass in the `struct db_stmt` since really they refer to the statement that was
executed and not the db.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
These are based on top of the basic column access functions, and act as a
small type-safe wrapper, that also does a bit of validation.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This has a slight side-effect of removing the actual begin and commit
statements from the `db_write` hooks, but they are mostly redundant anyway (no
harm in grouping pre-init statements into one transaction, and we know that
each post-init call is supposed to be wrapped anyway).
Signed-off-by: Christian Decker <decker.christian@gmail.com>
These are used to do one-time initializations and wait for pending statements
before closing.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
I was hoping to get rid of these by using "ON CONFLICT" upserts, however
sqlite3 only started supporting them in version 3.24.0 which is newer than
some of our deployment targets.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This is the first step towards being able to extract information from query
rows. Only the most basic types are exposed, the others will be built on top
of these primitives.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
For some of the query methods in the next step we need to have an idea of
whether the stmt was executed (db_step function) so let's track that
explicitly.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
These do not require the ability to iterate over the result, hence they can be
migrated already.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
These functions implement the lookup of the query, and the dispatch to the
DB-specific functions that do the actual heavy lifting.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This gets rid of the two parallel execution paths of read-only and write
queries, by explicitly stating with each query whether it is a read-only
query, we only need to remember the ones marked as write queries.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
All drivers will have to reach into it, so put it in a place that is reachable
from the drivers, along with all other definitions.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
This is the counterpart of the annotations we did in the last few commits. It
extracts queries, passes them through a driver-specific query rewriter and
dumps them into a driver-specific query-list, along with some metadata to
facilitate processing later on. The generated query list is then registered as
a `db_config` and will be loaded by the driver upon instantiation.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
We will soon generalize the DB, so directly reaching into the `struct db`
instance to talk to the sqlite3 connection is bad anyway. This increases
flexibility and allows us to tailor the actual implementation to the
underlying DB.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
These two simple macros have a twofold use:
1) They serve as annotations for the query extraction tool to find them when
extracting queries from the C source code.
2) They replace the actual queries with names that can be used to lookup the
queries in a table again, once they have been rewritten into the target SQL dialect.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
`db_select_prepare` was prepending the "SELECT" part in an attempt to limit
its use to read-only statements. This is leads to the queries in the code not
actually being well-formed, which we'll need in a later commit, and was also
resulting in extra allocations. This switches the behavior to just enforce a
"SELECT" prefix being present which allows us to have well-formed queries in
the code again and avoids the extra allocation.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
We need to have full DB queries that can be extracted at compile time later in
order to be able to rewrite them in other SQL dialects. In addition we had a
bit of unnecessary code-duplication in db_select and db_select_prepare. Now
the former uses the latter internally.
Signed-off-by: Christian Decker <decker.christian@gmail.com>
Rusty Russell
darosior
Vasil Dimov
Christian Decker
lisa neigut
Michael Schmoock
trueptolemy
Saibato