< achow101> luke-jr: it seems that vmbuilder is no longer working for me :(
< luke-jr> :|
< luke-jr> I made a base image by hand and can't get the gitian builds to work either
< luke-jr> hmm, so I got it to build finally, but the windows yml produces mismatches of the .tar.gz and .exe installers (the zips match)
< luke-jr> has anyone published these files for comparison?
< wumpus> luke-jr: uploading
< wumpus> luke-jr: if you made a base image by hand, though, can you make sure you have the correct version of NSIS installed? that's the primary cause of mismatches in the installer, I'd say
< luke-jr> wumpus: nsis packages seem to match
< wumpus> I've uploaded only the windows ones you mentioned, if you need any others let me know
< mess110> hi, can I get a few reviews for https://github.com/bitcoin/bitcoin/pull/13248 please?
< wumpus> #13248
< gribble> https://github.com/bitcoin/bitcoin/issues/13248 | [gui] Make proxy icon from statusbar clickable by mess110 · Pull Request #13248 · bitcoin/bitcoin · GitHub
< wumpus> looks like it has enough utACKs for a change like this, I'll give it a quick try
< Zenton> if I store a wallet with a strong passphrase and save that wallet dat file in an usb memory, is that security enough or do you recommend me to also encrypt all the usb memory?
< wumpus> Zenton: better ask in #bitcoin, this is a developer channel, you'd probably want to mention your perceived threat model for 'secure enough' as well
< Zenton> thanks wumpus
< provoostenator> Is it a new "feature" that you can't interrupt (ctrl + c, or even ctrl + z) gitian-build.py? I forgot to run it inside a screen :-)
< arubi> oh man and I on tmux, and I hit ctrl+c on the gitian pane instead of on the `tail -f build.log` one and thought some divine intervention happened that didn't kill the build
< cfields> wumpus: I didn't sign rc1 because I wasn't sure if the determinism issue was relevant. Now that we've tracked it down and it's Linux-only, would you like me to sign for rc1, or just skip and do an rc2?
< kevink> If this is true it looks like someone found a way to increase the blocksize via a soft-fork. It sounds very dangerous, I thought the Core developers might want to know https://twitter.com/StopAndDecrypt/status/1031619644627582976
< sipa> kevink: of course softforks to increase capacity are possible
< sipa> a softfork requires consensus to deploy, though
< sipa> and potential resource increases don't affect those who do not choose to follow them
< kevink> I asked Luke before and he said further blocksize increases would require a hard-fork. Thats why I thought it was a big deal. I guess not. 😂
< sipa> this is just research
< sipa> research about upgrade mechanisms is obviously a good thing
< kevink> Are they? You don't think this is a potential attack vector?
< sipa> no
< sipa> if it is an attack vector, it should be fixed
< sipa> so far i haven't seen anything to suggest that it is
< gmaxwell> It's a waste of everyones time to bother speculating about things that the people suggesting them won't disclose.
< kevink> Fair enough, we can talk about it after the presentation. ;)
< gmaxwell> as usual, everyone in that discussion is probably misunderstanding each other.
< gmaxwell> (as usual in the sense that it's twitter)
< StopAndDecrypt> i hold my conclusions until its presented
< sipa> thanks