Ensure that calling Next twice in a row without first calling Reset is
safe when the iterator is at the end of its list. Also alter the
towerListIterator to call Reset after hitting an error on Next.
This commit adds a test that shows that it is possible to cause the
AddressIterator to panic if the `Next` method is ever called twice when
the iterator is at the end of its list without Reset first being called.
In this commit, a new `ExhaustedSessionFilter` function is added and
used as a PostEvalFilterFn used when loading sessions from the DB. It
allows us to not unnecessarily load exhausted sessions into memory for
areas of the code where they will not be needed.
This commit adds a new PostEvaluateFilterFn call-back option to
`wtdb.ClientSessionListCfg`. This call-back is run _after_ all the other
evaluation call-backs in `wtdb.ClientSessionListCfg` and is only used to
determine if the session should be included in the returned list or not.
In this commit, a PreEvaluateFilterFn option is added to the
wtdb.ClientSessionListCfg and it is used instead of a separate
ClientSessionFilterFn parameter. This neatens quiet a few function
signatures.
This commit adds a deleteSessionFromTower method which can be used to
dial the tower that we created a given session with and then sends that
tower the DeleteSession method.
In this commit, a thread-safe min-heap is implemented. It will carry
sessionCloseItems which carry a sessionID and a block height at which
the session should be closed.
This commit adds a new ListClosableSessions method to the tower client
DB. This method will return a map of sessionIDs to block heights. The
IDs belong to sessions that are considered closable and the block
heights are the block height at which the last associated channel for
the session was closed in.
This commit adds a `MarkChannelClosed` method to the tower client DB.
This function can be called when a channel is closed and it will
check the channel's associated sessions to see if any of them are
"closable". Any closable sessions are added to a new
`cClosableSessionsBkt` bucket so that they can be evaluated in future.
Note that only the logic for this function is added in this commit and
it is not yet called.
In this commit, a migration is added that adds an index from channel to
sessionIDs (using the DB-assigned session IDs). This will make it easier
in future to know which sessions have updates for which channels.
In this commit, a new session-ID index is added to the tower client db
with the help of a migration. This index holds a mapping from a
db-assigned-ID (a uint64 encoded using BigSize encoding) to real session
ID (33 bytes). This mapping will help us save space in future when
persisting references to sessions.
In this commit, a new ClientSessionFilterFn parameter is added to the
DB's ListClientSession method which can be used to allow the caller to
specify a filter function for filtering sessions read from the DB.
Currently all filtering of sessions are done after the sessions have
been read from the DB, so adding this option should provide some
efficiency.
In this commit, a small migration is added to the watchtower client DB
to ensure that there is an entry in the towerID-to-sessionID index for
all towers in the db regardless of if they have sessions or not. This is
required as a follow up to migration 1 since that migration only created
entries in the index for towers that had associated sessions which would
lead to "tower not found" errors on start up.
With this commit we bump the github.com/btcd/btcec/v2 library to v2.3.2
which implements the MuSig2 BIP version v1.0.0rc2. With this the
github.com/btcsuite/btcd/btcec/v2/schnorr/musig2 package becomes
v1.0.0rc2 and the github.com/lightningnetwork/lnd/internal/musig2v040
stays at the old v0.4.0 version.
In this commit, a migration is done that takes all the AckedUpdates of
all sessions and stores them in the RangeIndex pattern instead and
deletes the session's old AckedUpdates bucket. All the logic in the code
is also updates in order to write and read from this new structure.
In this commit, the code for migration 4 is added. This migration takes
all the existing session acked updates and migrates them to be stored in
the RangeIndex form instead. Note that this migration is not activated
in this commit. This is done in a follow up commit in order to keep this
one smaller.
In this commit, we add the ability to add a wtdb version migration that
does not get given a transaction but rather a whole db object. This will
be useful for migrations that are best done in multiple transaction in
order to use less RAM.
In preparation for an upcoming commit where some helper functions will
need access to the ClientDB's ackedRangeIndex member, this commit
converts those helper functions into methods on the ClientDB struct.
Refactor the putClientSessionBody to take in a session sub-bucket rather
than the top-level session bucket. This is mainly to make an upcoming
commit diff easier to parse.
In this commit, a new concept called a RangeIndex is introduced. It
provides an efficient way to keep track of numbers added to a set by
keeping track of various ranges instead of individual numbers.
Notably, it also provides a way to map the contents & diffs applied to
the in memory RangeIndex (which uses a sorted array structure) to a
persisted KV structure.
In this commit, a new channel-ID index is added to the tower client db
with the help of a migration. This index holds a mapping from a
db-assigned-ID (a uint64 encoded using BigSize encoding) to real channel
ID (32 bytes). This mapping will help us save space in future when
persisting references to channels.
In this commit a migration of the tower client db is done. The migration
creates a new top-level cChanDetailsBkt bucket and for each channel
found in the old cChanSummaryBkt bucket, it creates a new sub-bucket. In
the subbucket, the ClientChanSummary is then stored under the
cChannelSummary key. The reason for this migration is that it will be
useful in future when we want to store more easily accessible data under
a specific client ID.
In this commit, the bug demonstrated in the previous commit is fixed.
The locking capabilities of the AddressIterator are used to lock
addresses if they are being used for session negotiation. So now, when a
request comes through to remove a tower address then a check is first
done to ensure that the address is not currently in use. If it is not,
then the request can go through.
In this commit, a new test is added to demonstrate that an error is
thrown if a user attempts to remove a tower during session negotiation
even if no current negotiation is taking place with the tower.
This commit upgrades the wtclient package to make use of the new
`AddressIterator`. It does so by first creating new `Tower` and
`ClientSession` types. The new `Tower` type has an `AddressIterator`
instead of a list of addresses. The `ClientSession` type contains a
`Tower`.
In this commit, a new AddressIterator type is added. It is a type that
can be used to iterate over a list of addresses. It strictly disallows
the list of addresses it holds to be empty. It also allows callers to
place locks on certain addresses in order to prevent other callers from
removing the addresses in question from the iterator.
In this commit, a test is added to demonstrate that if a client tries to
update the address of a tower for an active tower session, then this new
address will not be used until the client restarts. This is a bug that
will be fixed in a future commit.
This commit is a step towards prepping the watchtower client test
harness to be able to handle the case where the client connects to
multiple mock servers.
