1. 
   ryanofsky commented at 3:52 pm on February 5, 2020: member

   Filing this issue to record steps to reproduce a small regression from #17261, which is fixed in #18067. Reproducing the bug has to be done manually because we don’t have a framework for writing wallet backwards compatibility tests (#12134 is stuck in review)

   Steps to reproduce

    0MERGE=2d6e76af240969aa284cd4c3d376493988e218c2 # 17261 merge commit
    1PK1=0296b538e853519c726a2c91e61ec11600ae1390813a627c66fb8be7947be63c52 # key from wallet_address_types.py
    2PK2=037211a824f55b505228e4c3d5194c1fcfaa15a456abdf37f9b9d97a4040afc073 # key from wallet_address_types.py
    3
    4git checkout $MERGE^
    5make
    6src/bitcoind -regtest &
    7ADDRESS=$(src/bitcoin-cli -regtest addmultisigaddress 1 "[\"$PK1\", \"$PK2\"]" true legacy | jq -r .address)
    8src/bitcoin-cli -regtest stop
    9
   10git checkout $MERGE
   11make -C src bitcoind
   12src/bitcoind -regtest &
   13src/bitcoin-cli -regtest getaddressinfo $ADDRESS | jq .solvable # BUG: expect true, actual false
   14src/bitcoin-cli -regtest stop
   

   Expected behavior

   getaddressinfo reports multisig address is solvable

   Actual behavior

   getaddressinfo returns "solvable": false

2. ryanofsky added the label Bug on Feb 5, 2020
3. fanquake added the label Wallet on Feb 5, 2020
4. 
   Sjors commented at 7:05 pm on February 11, 2020: member

   I wrote a test on top of #12134 that reproduces this by calling addmultisigaddress in v0.19.0.1, loads that wallet on master and fails. #18067 fixes it: https://github.com/Sjors/bitcoin/commit/bacf55990905c0257c7bc45bbbd2b028716efc7f
5. 
   Empact commented at 1:45 am on February 19, 2020: member

   According to @achow101 this should be fixed now that #18067 is merged.
6. fanquake closed this on Feb 19, 2020

   ---

7. MarcoFalke locked this on Feb 15, 2022

Contributors
ryanofsky Sjors Empact

Labels
Bug Wallet