In this commit, we fix a bug that would cause the entire db to shutdown
if hit a panic (since db operations in the main buckets exit with a
panic) while executing a txn call back. This might be a postgres error
we need to check, so we don't want to bail out, and instead want to pass
up the error to the caller so we can retry if needed.
With the new postgres concurrency control, an error may come from a
bucket function that's actually a postgres error. In this case, we need
to return early so we can retry the txn. Otherwise, we'll be working
with an aborted tx, and never actually return the error so we don't auto
retry.
Some sub-systems like btcwallet will return an error from the database,
but they won't properly wrap it. As a result, we were unable to actually
catch the serialization errors in the first place. To work around this,
we'll now attempt to parse the error string directly.
In this commit, rather than attempt to reduce the scope of the current
`CODEOWNERS` file, we instead rename it as a way to stop Github from
automatically assigning reviewers based on it. Instead, we put things in
a "hint" mode as a way for PR OPs to figure out who they should manually
assign PRs today.
With the CODEOWNERS file reduced to a code review assignment hint, we
aim to instead promote usage of the `@code-review` team which will use a
load balancing algorithm for automatic code review assignment.
In this commit, we update new Taproot related TLVs (nonces, partial sig,
sig with nonce, etc). Along the way we were able to get rid of some
boiler plate, but most importantly, we're able to better protect against
API misuse (using a nonce that isn't initialized, etc) with the new
options API. In some areas this introduces a bit of extra boiler plate,
and where applicable I used some new helper functions to help cut down
on the noise.
Note to reviewers: this is done as a single commit, as changing the API
breaks all callers, so if we want things to compile it needs to be in a
wumbo commit.
In this commit, we add two new methods that simplify usage of an
handling a value wrapped in an Option. Thes methods allow a caller to
force the nil/None check up front, and either obtain the value, or fail
the current execution or test.
This should replace most usages of `UnsafeFromSome` where ever used
today.
These two messages will be used to implement the new and improved co-op
closing protocol. This PR also show cases how to use the new
`tlv.OptionalRecord` type to define and handle TLV level parsing.
I think we can make one additional helper function to clean up some of
the boiler plate for the encode/decode.
In this commit, we add a new `Zero` method for the `RecordT` type. This
method allows a caller to create the zero record for a type without
needing to reference the actual TLV type.
With this we go from this:
```go
sig1 := tlv.ZeroRecordT[tlv.TlvType1, Sig]()
```
To this:
```
sig1 := c.CloserNoClosee.Zero()
```
This builds on the concurrent queue to create a generic way to allow
goroutines to pub/sub information. An example includes being notified
each time a state machine is able to carry out a new state transition.
The error was never used as the init couldn't return an error, so we do
away with that. We also modify the main event loop dispatch to more
closely match other areas of the codebase.
In this commit, we make all calls to disconnect after a ping/pong
violation is detected in the `PingManager` async. We do this to avoid
circular waiting that may occur if the disconnect call back ends up
waiting on the peer goroutine to be torn down. If this happens, then the
peer goroutine will be blocked on the ping manager fully tearing down,
which is blocked on the peer disconnect succeeding.
This is a similar class of issue we've delt with recently as pertains to
the peer and the server: sync all back execution must not lead to
a circular waiting loop.
Fixes#8379
We add a WhenSome that'll pass in the actual underlying value, and not
just the record.
We also add `SomeRecord` to make it easier to create TLV records w/
`Some` set.
This will be useful when decoding optional TLV records, as we can use
this to look up in the `typeMap` for a given field to see if we decoded
it or not.
With this change, callers can now examine a struct instance of the
`TlvType` instance to figure out programmatically which integer type is
maps to. This'll be useful when decoding optional TLV types into a wire
struct (knowing when to set it to Some vs None).
In this commit, we modify the RecordT type to allow callers to re-use
the Record definition of a different type, but use the new type param to
override the integer type used on the wire.
This will let use do things like encode a signature using the same
RecordProducer instance, but with a diff type in another context.
The upcoming use for this is allowing our `lnwire.Sig` type to be
encoded in the same message using distinct TLV integer types (new co-op
close protocol).
In this commit, we add a new type, `RecordT[T, V]` to reduce some of the
common boiler plate for TLV types. This type lets you take either a
primitive type, or an existing Record, and gain common methods used to
create tlv streams.
It also serves as extra type annotation as well, since wire structs can
use this to wrap any existing type and gain the relevant record methods.
This implementation ensures that the very definition of the field also
binds the TLV type value. It does this by using the generated code to
map a struct like TlvType1 to an actually Type like Type(1).
This shouldn't need to be run again, as this implementation restricts
things to just values 0-99, due to a hard upper limit with the way Go
unions work under the hood.
In this commit, we add some new code generation to the codebase. As
we'll see in a future commit, this'll allow us to create a new Record[T,
V] type, where T is actually a concrete _struct_ that implements a
special interface that deems it as a valid TLV type.
In this commit, we make a new sub-module from the `fn` package. This
lets other projects use it w/o pulling in all of lnd, and also other
sub-packages within lnd.
We also add a temp replace directive for new module. We'll remove this
once we tag the final version.
We've only ever made macaroons with the v2 versions, so we should
explicitly reject those that aren't actually v2. We add a basic test
along the way, and also add a similar check for the version encoded in
the macaroon ID.
When we first go to boot up the syncer, when we're in the active phase,
after we do the historical sync, we'll send a timestamp message that we
want everything, then transition to the passive mode. The test didn't
account for this extra message before, as the last test was being
re-used here (ran in parallel).
We fix this by asserting that the first expected message is sent, then
also the follow up messages as well.
Olaoluwa Osuntokun