< bitcoin-git>
[bitcoin] domob1812 closed pull request #15259: Use real HTTP bind address in curl RPC help (master...decoupled-rpchelp) https://github.com/bitcoin/bitcoin/pull/15259
< stevenroose>
What are some "network identifiers" that can be used to generally refer to a "bitcoin network"? I.e. Mainnet, Testnet, Liquidv1, ...
< stevenroose>
I'm thinking the network magic would be a good idea.
< promag>
losing my mind with bitcoin-qt shutdown
< promag>
event loop should be the last man standing
< tryphe>
stevenroose, the network magic bytes for p2p: mainnet: f9beb4d9, testnet: 0b110907, regtest: fabfb5da
< dongcarl>
Hey all, I've been working on making bitcoin builds with Guix, which I think can replace gitian for our release process one day. It is a very exciting package manger which will bring much better supply chain security to Bitcoin. Let me know your thoughts! https://github.com/bitcoin/bitcoin/pull/15277
< sipa>
stevenroose: i don't understand the question
< promag>
"bitcoind -nowallet -wallet=w1" should fail?
< luke-jr>
promag: or perhaps void the bitcoin.conf wallets, and load wallet w1?
< promag>
luke-jr: -wallet=w1 -nowallet -wallet=w2 -> only w2 is loaded?
< luke-jr>
promag: I guess
< luke-jr>
I suppose it makes sense if a user has an alias
< gleb>
I've modified transaction transmission and sometimes (very rarely though) I'm getting "non-mandatory-script-verify-flag (Witness program hash mismatch)". Any idea how it might happen or where should I look for? Transactions are generated by "bitcoin-cli sendtoaddress"
< gleb>
This error pops up after one of the nodes receives a transaction through p2p and attempts to verify it (not the first node which crafts it)
< stevenroose>
sipa: like I'm writing a BIP and I need a value to refer to what network something is relevant on (f.e. say the PSBT prefix (/HRP)) and instead of inventing a new value, I'm looking to re-use an existing one.
< stevenroose>
I could do network magic bytes, psbt prefix, BIP44 coin type id, base58check address prefix byte (f.e. fixed at p2pkh), ... Address-specific ones are probably very bad.
< stevenroose>
I think magic bytes are quite solid. More available in implementations than BIP44 coin type ids, f.e..
< sipa>
stevenroose: if it's a BIP, no need to refer to anything but bitcoin and testnet
< sipa>
you can have some other repository with ids or so
< luke-jr>
stevenroose: BIP70 had chain ids as strings