To give the CLI user the option to delete a single or multiple payments
in one command, we expose the DeletePayment and DeleteAllPayments
RPCs to the command line as well. Due to the similarity of the two RPCs
we combine them into a single command with multiple flags.
To make the command a bit more safe to run without arguments, we
consciously switch the logic of the RPC flag "failed_payments_only"
which is false by default into a "--include_non_failed" in the CLI which
is false by thefault. So a user running the command without knowing what
they are doing are only deleting failed payments by default, not all of
the payments.
The use of golang.org/x/crypto/ssh/terminal was marked as deprecated
since that code was moved to its own repo golang.org/x/term.
With this commit we use the new package name directly, which was already
pulled in under its new name anyway.
Having one file per sub command seems a bit too excessive and isn't
really implemented for any of the other commands. So we just group the
Mission Control commands into their own file for now.
To make the "Payments" category a bit less overloaded and to move the
Mission Control configuration commands away from the "root" category, we
create the new "Mission Control" category that we move the commands
into. We do this in a single commit so the next one where we move them
into the same file can be a pure code move without any additional
changes.
With this commit we move all commands that can be found within the
"Payments" category into its own file called cmd_payments.go. The only
exception are the Mission Control related commands which we are going to
move to their own category and file in the next commit.
In this commit, a new method `cleanClosedChannels` is added and called
when a circuit map is created. This method will delete the payment
circuits and keystones for closed channels.
It was being considered a misbehaviour from the intermediate hop which
penalized that hop, which is a bit harsh considering mobile nodes. We
change it be considered as other channel update types and only penalize
the channel that failed.
With the middleware handler in place, we now need to add a new gRPC
interceptor to the interceptor chain that will send messages to the
registered middlewares for each event that could be of interest to them.
With this commit we introduce the concept of RPC middleware: A mechanism
similar to the existing channel or HTLC interceptors but this time for
gRPC messages themselves.
An RPC middleware can register itself to the main RPC server to get
notified each time a new gRPC request comes in, a gRPC response is sent
back or a streaming RPC is connected. The middleware can
validate/inspect incoming requests and modify/overwrite outgoing
responses.
Since this also opens the door for malicious software to interfere with
lnd in a negative way, we bind everything to macaroons with custom
caveat conditions: A middleware declares upon registration which custom
caveat name it can handle. Only client requests that send a macaroon
with that custom caveat will then be given to the middleware for
inspection. The only exception is if the middleware instead registers
to use the read-only mode. In that mode it will be able to intercept
all requests/responses, even those not made with a special encumbered
macaroon. But the middleware won't be able to alter responses in the
read-only mode. Therefore requests with the default, unencumbered macaroons
can never be modified by any middleware.
The custom RPC middleware logic that we are going to add in the next
commits will need to log under their own sub logger so we add one with a
new subsystem name.
With the new condition and checker in place, we can give the end user
the ability to add such a custom caveat to a baked macaroon.
There won't be an RPC counterpart for this operation since all first party
caveats currently are only added on the client side.
