ArctVaulMarsHMPJ has quit [Read error: Connection reset by peer]
ArctVaulMarsHMPJ has joined #bitcoin-core-dev
Henrik has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
Henrik has joined #bitcoin-core-dev
yanmaani has joined #bitcoin-core-dev
Henrik has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
bitcoin-git has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] hebasto opened pull request #22833: ci: Switch to Previous Visual Studio 2019 image to avoid failure (master...210830-appveyor) https://github.com/bitcoin/bitcoin/pull/22833
bitcoin-git has left #bitcoin-core-dev [#bitcoin-core-dev]
jungly_ has joined #bitcoin-core-dev
aitorjs has joined #bitcoin-core-dev
nathanael has quit [Quit: connection reset by purr]
nathanael has joined #bitcoin-core-dev
Chris_Stewart_5 has joined #bitcoin-core-dev
<vasild>
If -onlynet=ipv6 is given, we would still connect to ipv4 seeds.
<laanwj>
vasild: yes, the scope of onlynet is *only* outgoing P2P connections
<vasild>
well, those are still outgoing p2p connections (short lived)
<laanwj>
the appveyor warning isn't exactly illuminating as to what is happening
<laanwj>
vasild: true, but they're made to names, not to addresses
<laanwj>
vasild: when using a proxy, bitcoind has no idea whether it's connecting to ipv4, ipv6 or something else completely :)
<vasild>
true
<laanwj>
this is the story behind the "name proxy", i don't know if that's still always a thing
<vasild>
but if connecting directly and there is a bad spy looking at my ipv4 outgoing traffic they will learn I run bitcoin, which I may have tried to avoid by using -onlynet=ipv6
<laanwj>
if it fails there's always the hardcoded IP seeds too
<laanwj>
in that case you shouldn't be using DNS at all, DNS leaks are the primary thing spies look at i think
<laanwj>
once you've looked up the DNS seed i doesn't really matter if you try to connect to it
<vasild>
right
<laanwj>
maybe it would have been better not to do the connect-to-name P2P thing, but always go with hardcoded seeds instead
<laanwj>
then again the DNS seeds are bound to be more up to date than hardcoded seeds so it's definitely a compromise
<laanwj>
there is already an option to use the DNS seeds and IIRC it's automatically disabled by some combinations of options, maybe it should include onlynet
<laanwj>
when connecting blindly to a name you simply cannot give a network guarantee
<laanwj>
and the whole point of that way of using the DNS seeds is that direct DNS lookups aren't possible (so behind a proxy)
klementtan has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
roconnor has joined #bitcoin-core-dev
<laanwj>
so i think the following parameter interaction would make sense: if -onlynet is provided, set -dnsseed=0
<laanwj>
i don't think it should be part of #22651's scope though
<bitcoin-git>
[bitcoin] hebasto closed pull request #22833: ci: Switch to Previous Visual Studio 2019 image to avoid failure (master...210830-appveyor) https://github.com/bitcoin/bitcoin/pull/22833
bitcoin-git has left #bitcoin-core-dev [#bitcoin-core-dev]
<laanwj>
the appveyor error on #22833 is amusingly unilluminating "EXEC : error : Building package openssl:x64-windows-static failed with: BUILD_FAILED"
<laanwj>
not sure where it gets the requirement from, i don't see it anywhere in our repo's msvc or appveyor metadata
<MarcoFalke>
same
<MarcoFalke>
Maybe some implict dependency for something else?
<laanwj>
i suppose so, it's a very common dependency
<hebasto>
the weirdest thing is that appveyor works fine on personal account
bitdex_ has quit [Quit: = ""]
Guyver2 has quit [Quit: Going offline, see ya! (www.adiirc.com)]
smartin has joined #bitcoin-core-dev
<MarcoFalke>
I could try clearing the cache ...
<laanwj>
right, it's unlikely to be a problem with openssl, it seems the compiler is simply missing... or at least not in the place the build script expects
SpellChecker has quit [Remote host closed the connection]
<MarcoFalke>
looks like appveyor fixed itself (I didn't do nothing)
roconnor has quit [Ping timeout: 256 seconds]
smartin has quit [Ping timeout: 244 seconds]
smartin has joined #bitcoin-core-dev
<laanwj>
ryanofsky: added
<laanwj>
MarcoFalke: fantastic
meshcollider has quit [Remote host closed the connection]
bitdex has joined #bitcoin-core-dev
<laanwj>
at least it fixed itself as mysteriously as it broke itself
goatpig has joined #bitcoin-core-dev
Guyver2 has joined #bitcoin-core-dev
jarthur has joined #bitcoin-core-dev
<laanwj>
is there a straightforward way to run the functional tests against a set of downloaded binaries? the distribution includes test_bitcoin but i wonder if i can run the functional ones somehow
meshcollider has joined #bitcoin-core-dev
aitorjs has joined #bitcoin-core-dev
roconnor has joined #bitcoin-core-dev
<MarcoFalke>
laanwj: Easiest would be to checkout the source at the exact version, then `./configure`, then move the binaries to $builddir/bitcoind etc
<MarcoFalke>
If you don't want to use configure, you can also use "sed" at your own risk
<josibake>
laanwj: you also set env variables for BITCOIND and BITCOINCLI pointing to the binaries you want to use before running ./test/functional/test_runner.py
<laanwj>
i was trying to create my own config.ini then set builddir appropriately, then move bitcoind and the tools there to mimic the build dir hierarchy, but "./test_runner.py --configfile /path/to/config.ini" doesn't work
AaronvanW has joined #bitcoin-core-dev
<laanwj>
it interprets the argument as test specification instead of passing it through to the individual tests
<josibake>
s/you also set/you can also set/
<laanwj>
looks like test_runner has the path to config.ini hardcoded anyway, so that wasn't going to work
<laanwj>
josibake: that seems useful!
<laanwj>
the functional tests only use those two?
<MarcoFalke>
They also use bitcoin-wallet
<MarcoFalke>
The bitcoin-util tests use bitcoin-util and bitcoin-cli
<laanwj>
MarcoFalke: thanks!
<laanwj>
looks like there is no environment var to override bitcoin-wallet path, anyhow i've symlinked things into a fake "src" hierarchy already anyway
<laanwj>
symlink test/functional/test_runner.py into it too, create your own test/config.ini in the same hierarchy, and it works :)
<MarcoFalke>
Maybe time to create a zip with the tests+ini in guix?
<laanwj>
that would make it slightly easier (though it's, pretty straightforward to do manually, just a matter of knowing what to do)
<midnight>
Uh. I'm sorry, I keep forgetting. You guys would prefer combined commits for gitian.sigs for when I do 6+ of them, or do you prefer one-per for the top-line summary?
<midnight>
combined so you don't have to merge prepeatedly, right?
aitorjs has quit [Ping timeout: 244 seconds]
<sipa>
midnight: i don't think it matters
lkqwejhhgasdjhgn has quit [Quit: Konversation terminated!]
<laanwj>
great, all the tests pass for risc-v 22.0rc3
<laanwj>
midnight: most people do one PR for non-codesigned one for codesigned, the number of individual commits in them doesn't matter
tripleslash has quit [Remote host closed the connection]
Talkless has joined #bitcoin-core-dev
davterra has joined #bitcoin-core-dev
vasild has quit [Remote host closed the connection]
vasild has joined #bitcoin-core-dev
aitorjs has quit [Ping timeout: 245 seconds]
<roconnor>
achow101: If GenericTransactionSignatureChecker requires PrecomputedTransactionData, is it still possible for data to be missing for the MissingDataBehavior to have effect?
<achow101>
roconnor: yes, PrecomputedTransactionData may be uninitialized
<achow101>
or rather not have anything precomputed
<achow101>
(uninitialized as in not having called PrecomputedTransactionData::Init, not c++ uninitialized)
lightlike has joined #bitcoin-core-dev
bitcoin-git has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] sipa opened pull request #22837: doc: mention bech32m/BIP350 in doc/descriptors.md (master...202108_bip350descdoc) https://github.com/bitcoin/bitcoin/pull/22837
bitcoin-git has left #bitcoin-core-dev [#bitcoin-core-dev]
davterra has quit [Quit: Leaving]
test has joined #bitcoin-core-dev
test has quit [Remote host closed the connection]
test_login has joined #bitcoin-core-dev
davterra has joined #bitcoin-core-dev
test_login has quit [Remote host closed the connection]
gnaf has joined #bitcoin-core-dev
baldur has quit [Ping timeout: 248 seconds]
gnaf has left #bitcoin-core-dev [#bitcoin-core-dev]
gnaf has joined #bitcoin-core-dev
Talkless has quit [Quit: Konversation terminated!]
baldur has joined #bitcoin-core-dev
bitcoin-git has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] achow101 opened pull request #22838: descriptors: Be able to specify change and receiving in a single descriptor string (master...multipath-descs) https://github.com/bitcoin/bitcoin/pull/22838
bitcoin-git has left #bitcoin-core-dev [#bitcoin-core-dev]
smartin has quit [Quit: smartin]
gnaf has quit [Remote host closed the connection]
fch has joined #bitcoin-core-dev
fch has quit [Quit: Leaving]
fch has joined #bitcoin-core-dev
fch has quit [Client Quit]
Guyver2 has quit [Quit: Going offline, see ya! (www.adiirc.com)]