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
1.) Add idle handler for cleaning pending ntzpool. There should not be two pending ntz's in the pool at a given time. Should be simple enough to write in some extra rules to the refresh or update_pool RPC calls. Done : 2 Nov
2.) Existing idle handler needs to properly fetch pending ntz from the pool w/ metadata rather than relaying a NOTIFY_REQUEST_NTZ_SIG unconditionally.
Done: 30 Oct
3.) Make sure thatinvalid signatureerror has been cleared up. (follow-up on e15fecc)
- Also clean up the existing res.tx_metadata_list response that is fed into a request_ntz_sig wallet call. We really shouldn't need to handle the metadata list as a list, and should be able to change this to a string. It is one contiguous hex blob currently, so a list with one element.
Done: 30 Oct
4.) Make sure that what is being done with checking the identity element & viewkeys is acceptable means for verifying sender is a notary node.This might not be a sufficient check for integrity due to the surjective nature of the function, and problem could arise due to being 'right-invertible'. This is a problem, if I am understanding correctly, since we could recover using a pre-image ... unsure of implications here EDIT: Probably only the case with deterministic viewkeys.
5.) Pending-tx-specific hash needs stored for database retrieval of the ptx_blob. Added a function in notary_server.cpp to calculate the hash and pair it with the ptx_blob. Rest of functions in flow from wallet -> rpc -> core -> blockchain -> db need modified to pass the hash through with the blob. Done: Oct 31
6.)Bring README.md up to speed (Mostly Done: 3 Nov), and add code flow document, for easier comprehension. Should go a long way in making things easier for those in KMD/ZEC ecosystem, in particular.
7.) Ensure that check against notary pubkeys also occurs daemon-side and not solely on wallet side. Need to look into further, but pretty sure that we currently lack verification in daemon. Maybe consider adding to blockchain as well.
The text was updated successfully, but these errors were encountered:
1.) Add idle handler for cleaning pending ntzpool. There should not be two pending ntz's in the pool at a given time. Should be simple enough to write in some extra rules to theDone : 2 Novrefresh
orupdate_pool
RPC calls.2.) Existing idle handler needs to properly fetch pending ntz from the pool w/ metadata rather than relaying aNOTIFY_REQUEST_NTZ_SIG
unconditionally.Done: 30 Oct
3.) Make sure thatinvalid signature
error has been cleared up. (follow-up on e15fecc)-
Also clean up the existingres.tx_metadata_list
response that is fed into arequest_ntz_sig
wallet call. We really shouldn't need to handle the metadata list as a list, and should be able to change this to a string. It is one contiguous hex blob currently, so a list with one element.Done: 30 Oct
4.) Make sure that what is being done with checking the identity element & viewkeys is acceptable means for verifying sender is a notary node.
This might not be a sufficient check for integrity due to the surjective nature of the function, and problem could arise due to being 'right-invertible'. This is a problem, if I am understanding correctly, since we could recover using a pre-image ... unsure of implications hereEDIT: Probably only the case with deterministic viewkeys.5.) Pending-tx-specific hash needs stored for database retrieval of theDone: Oct 31ptx_blob
. Added a function innotary_server.cpp
to calculate the hash and pair it with theptx_blob
. Rest of functions in flow fromwallet
->rpc
->core
->blockchain
->db
need modified to pass the hash through with the blob.6.)
BringREADME.mdup to speed(Mostly Done: 3 Nov), and add code flow document, for easier comprehension. Should go a long way in making things easier for those in KMD/ZEC ecosystem, in particular.7.) Ensure that check against notary pubkeys also occurs daemon-side and not solely on wallet side. Need to look into further, but pretty sure that we currently lack verification in daemon. Maybe consider adding to blockchain as well.The text was updated successfully, but these errors were encountered: