< kallewoof>
ossifrage harding wumpus: I am seeing a significant increase in 'prev block not found' on my end. It's really frequent these days. (running on master)
< kallewoof>
I hadn't seen it a single time as of a few months ago. I see it almost every time I glance at the console output now.
< achow101>
kallewoof: is it possible that you have connected to some bcash nodes or nodes from one of the billion forks?
< kallewoof>
achow101: I haven't really done anything specific to the node. It's been running for some time though.
< kallewoof>
I might tweak it to show me more details when that error happens though.
< achow101>
check your peers list?
< achow101>
if you enable debug=net, you should also get more info as to who sent you the bad block
< kallewoof>
Ah good point
< provoostenator>
kallewoof: I've noticed this at well on my crappy-ARM-node, so didn't think much of it. Will do the debug=net thing.
< jonasschnelli>
sipa: sry for the delay on your descriptors PR review... I'm back at my desk end of the week and will continue with review and tests
< Fuzzbawls>
Hey, just wanted to pop in and say that I think I've done about all I can with #13732 without going outside the `depends/` path for file edits
< kallewoof>
provoostenator: I don't see "prev block not found" in that log tho. Different issue?
< provoostenator>
That happens directly after it received headres from that peer
< provoostenator>
But that doesn't show up in my 'grep "peer=167" that I used for the gist.
< kallewoof>
ahh i see
< provoostenator>
(and the next line is where it says "Mishaving: .... peer=167"
< Fuzzbawls>
provoostenator: looks to me like that peer is rejecting every transaction or inv you're sending it, while at the same time requesting data for transactions that you don't have.
< provoostenator>
Shouldn't normally behaving peers request some headers, decide they're not on the same chain and then leave?
< Fuzzbawls>
could be a client that never updated/changed the message header to suit whatever network it is supposed to be running on
< Fuzzbawls>
like a pre-beta BCH client perhaps?
< provoostenator>
BCH transactions would show up on Google, and that height is much too low (way before the fork).
< Fuzzbawls>
thats true
< Fuzzbawls>
maybe it's just some client that has it's head up its rear for whatever reason and cannot recognize a difference in consensus
< Fuzzbawls>
it's advertising 0.14.99...not exactly "recent" (nor from any release tag...so could be running with some out-of-whack modifications)
< bitcoin-git>
[bitcoin] Sjors opened pull request #13750: depends: use MacOS friendly sed syntax in qt.mk (master...2018/07/depends-qt-macos) https://github.com/bitcoin/bitcoin/pull/13750
< provoostenator>
Saw two more of them, or the same on different IP's. Same pattern.
< bitcoin-git>
[bitcoin] 251Labs opened pull request #13751: Utils and libraries: Drops the boost/algorithm/string/split.hpp dependency (master...patch/remove_boost_split) https://github.com/bitcoin/bitcoin/pull/13751
< timothy>
uhm, I see that commit 634bd970013eca90f4b4c1f9044eec8c97ba62c2 broked compatibility with python2. is it known?
< achow101>
timothy: python2 compatibility was dropped a while ago
< timothy>
achow101: why? centos7/rhel7 doesn't have python2 by default
< timothy>
and so you need to use external/unsupported repositories that is bad for security
< timothy>
s/python2/python3/
< achow101>
because there's features in python3 that we want to use that aren't available in python2
< achow101>
it also makes for easier to maintain code as you don't always have to make the code work across both py2 and 3
< bitcoin-git>
[bitcoin] laanwj closed pull request #12257: [wallet] Use destination groups instead of coins in coin select (master...feature-addrgrouped-coinselect) https://github.com/bitcoin/bitcoin/pull/12257
< luke-jr>
[13:45:17] <timothy> achow101: why? centos7/rhel7 doesn't have python2 by default <-- ouch
< luke-jr>
anyway, we don't need Python to build, right? just for tests?
< timothy>
luke-jr: python3, I corrected myself in the next message
< luke-jr>
timothy: yes, still ouch :P
< timothy>
so probably it's a good thing to keep python2 compatibility
< luke-jr>
dunno, I think that ship has sailed
< luke-jr>
I won't object if someone wants to do the work, but I think there's been explicit efforts to update to Py3 stuff that's not compatible
< luke-jr>
Python 3 is nearly a decade old at this point. Seems pretty unreasonable for RHEL 7 to have omitted it. :/
< bitcoin-git>
[bitcoin] kallewoof reopened pull request #10386: [WIP] [wallet] Optional '-avoidreuse' flag which defaults to not reusing addresses in sends (master...feature-white-black-address) https://github.com/bitcoin/bitcoin/pull/10386
< promag>
provoostenator: there are tabs in the touched lines in src/crypto/sha256_sse4.cpp
< promag>
should I skip that file?
< promag>
looks like it's an exception
< provoostenator>
There's also a bunch of accidental whitespace in that file.
< provoostenator>
So maybe fix that file manually rather than with scripted diff? Or indeed skip it.
< promag>
provoostenator: skip for now, I'll see if it's the only case
< promag>
I could s/\t/ /g
< promag>
hmm vm, that wouldn't work
< promag>
*nevermind
< promag>
provoostenator: indeed it's the only case, I'll skip it and let the PR be just scripted diff
< wumpus>
yes, python2 support has been dropped, it would be unreasonable to bring it back now
< wumpus>
ask your distribution to upgrade their version of python
< wumpus>
or use the distributed binaries, or cross-build the executables, as for embedded targets--you only need python for build and tests, not to run the results
< cfields>
wumpus: I believe depends builds rely on python in a few places. Possibly only for the X dependencies.
< cfields>
though it's never caused any trouble that I'm aware of, so I assume they're python2/3 compat
< cfields>
oh, sorry. "only for build and tests". You had that covered :)
< achow101>
petertodd: your testnet dns seeder appears to be dead
< bitcoin-git>
bitcoin/master 14788fb Chun Kuan Lee: [travis] Don't store debug info if --enable-debug is set
< bitcoin-git>
bitcoin/master 7a9bca6 MarcoFalke: Merge #13650: [travis] Don't store debug info if --enable-debug is set...
< bitcoin-git>
[bitcoin] MarcoFalke closed pull request #13650: [travis] Don't store debug info if --enable-debug is set (master...travis-debug-fix) https://github.com/bitcoin/bitcoin/pull/13650