You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When a node is offline, it may start to accumulate /blockchain
lag/. Blockchain lag is defined as the number of blocks that are in
the blockchain, but that the node has not yet seen and processed.
A large blockchain lag can cause a number of symptoms:
Failing payments with WIRE_UNKNOWN_INVOICE_DETAILS / 16399
despite the invoice definitely coming from the recipient: the CLTV
is too low for the recipient to accept it, hence it sends back a
signal that some choice on the sender was wrong. This can be a
disagreement on payment rules, or in this case a disagreement on
what is far enough in the future for the payment to be safe.
Missing incoming on-chain payment: we will see them and add them
once we process the block that contained the transaction.
Missing confirmations for sent transactions, and missing spent
marking of outputs used in a sent transaction: both of these will
be set when we witness the transaction in a block. Before that you
may refer to the reserved flag in listfunds to determine if an
output is likely soon to be marked as spent and should therefore
not be used. This reflects the intrinsic issue with confirmations
on blockchains, since there is this limbo state between broadcast
and confirmation, and so we chose to expose these details rather
than pretending that broadcast is confirming.
For all of these the cause is the same, and addressing it,
automatically addresses them too. As such we are working on improving
the blockchain lag, by synchronizing in the background (not counted
towards the node's runtime for accounting!), and to optimize the sync
of the last couple of blocks for interactive (user) sessions.
The text was updated successfully, but these errors were encountered:
The blockchain lag issues have been resolved on the server side. With the inclusion of ElementsProject/lightning#7190, the problems related to small blockchain lag have also been mostly addressed. Closing.
When a node is offline, it may start to accumulate /blockchain
lag/. Blockchain lag is defined as the number of blocks that are in
the blockchain, but that the node has not yet seen and processed.
A large blockchain lag can cause a number of symptoms:
WIRE_UNKNOWN_INVOICE_DETAILS
/ 16399despite the invoice definitely coming from the recipient: the CLTV
is too low for the recipient to accept it, hence it sends back a
signal that some choice on the sender was wrong. This can be a
disagreement on payment rules, or in this case a disagreement on
what is far enough in the future for the payment to be safe.
once we process the block that contained the transaction.
spent
marking of outputs used in a sent transaction: both of these will
be set when we witness the transaction in a block. Before that you
may refer to the
reserved
flag inlistfunds
to determine if anoutput is likely soon to be marked as
spent
and should thereforenot be used. This reflects the intrinsic issue with confirmations
on blockchains, since there is this limbo state between broadcast
and confirmation, and so we chose to expose these details rather
than pretending that broadcast is confirming.
For all of these the cause is the same, and addressing it,
automatically addresses them too. As such we are working on improving
the blockchain lag, by synchronizing in the background (not counted
towards the node's runtime for accounting!), and to optimize the sync
of the last couple of blocks for interactive (user) sessions.
The text was updated successfully, but these errors were encountered: