hsmd, splicing: wrong remote per-commit point provided when signing splice remote tx

[devrandom]

Issue and Steps to Reproduce

We (VLS) are seeing remote per-commitment point number 1 provided to signer instead of point number 0. This occurs when signing the remote commitment transaction on top of the new splice.

We have a vlsd log excerpt in https://gitlab.com/lightning-signer/vls-hsmd/-/merge_requests/132#note_1649538928 with the captured hsmd protocol messages.

I believe this happens when running any of the CLN test_splicing.py system tests.

[ddustin]

The commitment point behavior is changed by this PR: #6840

Prior behavior would be to do a complete commitment cycle (on original commitment and again for each splice RBF candidate) -- this PR changes the behavior to just do a single commitment for the new splice using the current commitment point number.

[ksedgwic]

Our testing is using #6840

[ddustin]

@ksedgwic Is this issue still happening?

[ksedgwic]

Apologies, not sure if still happening

…

On Thu, Jul 11, 2024 at 11:57 AM Dusty Daemon ***@***.***> wrote: @ksedgwic <https://github.com/ksedgwic> Is this issue still happening? — Reply to this email directly, view it on GitHub <#6870 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAJNPTFXLEZUXGLMUZLYALZL3IRVAVCNFSM6AAAAABKXRVQBSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMRTGY3TIMJSHA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Ken Sedgwick Bonsai Software, Inc. http://www.bonsai.com/ken/ (510) 269-7334 ***@***.*** Public Key: http://www.bonsai.com/ken/ken.asc GPG Fingerprint: 4695 E5B8 F781 BF85 4326 9639 BBFC E515 8602 5550