This change was largely motivated by an increase in high disk usage as a
result of channel update spam. With an in memory graph, this would've
gone mostly undetected except for the increased bandwidth usage, which
this doesn't aim to solve yet. To minimize the effects to disks, we
begin to rate limit channel updates in two ways. Keep alive updates,
those which only increase their timestamps to signal liveliness, are now
limited to one per lnd's rebroadcast interval (current default of 24H).
Non keep alive updates are now limited to one per block per direction.
The channeldb implements the persistent storage engine for lnd and
generically a data storage layer for the required state within the Lightning
Network. The backing storage engine is
boltdb, an embedded pure-go key-value store
based off of LMDB.
The package implements an object-oriented storage model with queries and
mutations flowing through a particular object instance rather than the database
itself. The storage implemented by the objects includes: open channels, past
commitment revocation states, the channel graph which includes authenticated
node and channel announcements, outgoing payments, and invoices
Installation and Updating
$ go get -u github.com/lightningnetwork/lnd/channeldb
791ba3eb50
