Transcript for #bitcoin-dev 2017/10/04

03:32 alexwaters is there current documentation for the bitcoind rpc api?
03:32 alexwaters er up-to-date
03:54 alexwaters oh nvm rpc has help command
15:22 waxwing Bit of a forlorn question, anyone got a clue why 53b8e28a5f2bddb167a7f7a3685808ff3101fc21b737349e4a950933fb36a8c9 is not confirming (been 3 hours now, about 9 blocks or so)?
15:24 arubi ah testnet, weird but my node doesn't have it
15:24 waxwing interesting. it's on smartbits.
15:25 waxwing i'm looking for segwit txs in the last block and i can't find any
15:26 arubi do wtxid commitments exist at all?
15:26 arubi (let's see)
15:27 arubi seems like that recent block don't have a commitment even, so probably not mining segwit on purpose
15:27 arubi some do but are empty
15:35 waxwing arubi, ah ok, of course that'd be a better way to check :)
15:35 waxwing yeah it's kind of a worry that no one apart from me is complaining ..? or has everyone moved to a secret testnet4 and not told me?
15:35 arubi hehe you know regtest is where all the fun happens
15:37 arubi but really I agree. testnet is a mess. I'm still running the node but at this point it's only taking up resources from me :(
15:37 waxwing for some things it's just much better to have a real network
15:38 waxwing not least because you actually want to do stuff with other people :) (in this case was just trying out a lightning node funding, oh well!)
15:38 waxwing more like tortoise network at this rate
15:39 arubi maybe a secret testnet isn't a bad idea
16:59 waxwing arubi, confirmed, only took 5 hours! what is this, bcash?
17:00 arubi obviously it's better than bcash because the chain is longer, making it the real bitcoin
22:48 vrs I run bitcoin on debian testing and when I try to start either of bitcoind or bitcoin-qt, I get an error along the lines of bitcoind: symbol lookup error: bitcoind: undefined symbol: _ZNK8UniValue7findKeyERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
22:48 vrs is this a known bug? it feels like it should affect more people
22:50 vrs I've already looked for UniValue and it seems to be in bitcoin's own codebase, not something like which was a library problem
22:56 vrs strace -e trace=file tells me it's loading a bunch of libraries, but the last lib is different depending on the binary I call, while the error is mostly the same
22:59 vrs are the errors I get
23:05 vrs ah, I was using bitcoin from unstable while my system is testing
23:07 vrs but how do I get a proper bitcoind on debian testing then
23:13 esotericnonsense vrs: do the binaries from work? you could try compiling from source otherwise
23:13 vrs this is specifically what I want to avoid when using debian
23:13 esotericnonsense haha
23:13 esotericnonsense this is specifically the reason that I stopped using debian
23:13 esotericnonsense :P
23:14 vrs I mean it's not that big a deal, I have other compiled stuff running, but the fewer packages I have to update manually, the better
23:14 esotericnonsense (because getting a reasonably updated version of everything entailed either using unstable or compiling)
23:15 esotericnonsense i think bitcoin is a unique case because updates should require more manual involvement than with other software anyway
23:15 esotericnonsense (i'm against autoupdate mechanisms for bitcoin)
23:15 vrs I'm a casual user, it just serves as a backend to armory in my case
23:16 esotericnonsense there's a PPA that is maintained for ubuntu, is there a way to use PPA's with debian? (it's been probably 5+ years since I last used deb)
23:16 vrs yes I can just add it but that's icky
23:17 esotericnonsense icky in what way? you want the latest upstream rather than what debian is currently packaging, seems like a reasonable way to do it
23:17 vrs it's mixing distros
23:17 esotericnonsense hehehe
23:18 vrs it's Bad, it's why you get symbol lookup errors like the one I did
23:18 esotericnonsense yeah
23:18 esotericnonsense if they don't provide a stable backport and others don't work, i don't see a way around compiling it