< GitHub30> [bitcoin] morcos opened pull request #6931: Skip BIP 30 verification where not necessary (master...skipBIP30) https://github.com/bitcoin/bitcoin/pull/6931
< GitHub1> [bitcoin] morcos opened pull request #6932: ModifyNewCoins saves database lookups (master...newCoinsThinAir) https://github.com/bitcoin/bitcoin/pull/6932
< midnightmagic> Is there a patch which ignores tx its already seen or is already in the mempool (preferrably pre-BlueMatt's mutation mech?)
< midnightmagic> or maybe lets me ban that host+src port peer for sending dupes even though it looks as though I have so far never seen a dupe sent?
< btcdrak> bfxda
< dcousens> hmph, not sure why, but since I re-compiled this node from master and started a re-download, its still ~4 months behind, slowly getting there, but taking forever
< dcousens> usually a full-resync/download only takes 2-3 days
< dcousens> This is over a week now
< dcousens> Any speculation as to why that might have changed since 0.11?
< sipa> is this a reindex or a redownload?
< dcousens> redownload
< dcousens> Would have been a re-index, but cancelled midway causing it to corrupt
< dcousens> not like its running at 100% either
< dcousens> plenty of resources all round
< dcousens> hmmm, wait, might have a theory, nvm :)
< Luke-Jr> dcousens: huh? nothing you can do should ever corrupt to the point reindex fails.
< dcousens> Luke-Jr: I must be doing something wrong then, restarted bitcoind, hashPrevBlock == view.GetBestBlock()
< dcousens> assert threw
< dcousens> sigh, any ideas for reconciliation?
< Luke-Jr> reindex is the usual solution for that..
< dcousens> Luke-Jr: we'll see how we go
< dcousens> interesting
< dcousens> Luke-Jr: since I started the re-index, my node is now showing up correctly on the network in a few places
< dcousens> I wonder if something was just borked that meant I only connected to 1 or 2 nodes?
< GitHub83> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/8fe30fb4d130...42f339ef780b
< GitHub83> bitcoin/master ff2a2af Wladimir J. van der Laan: build: don't distribute tests_config.py...
< GitHub83> bitcoin/master 42f339e Wladimir J. van der Laan: Merge pull request #6930...
< GitHub131> [bitcoin] laanwj closed pull request #6930: build: don't distribute tests_config.py (master...2015_11_testsconfig_nodist) https://github.com/bitcoin/bitcoin/pull/6930
< wumpus> dcousens: "hashPrevBlock == view.GetBestBlock()" is the error that you will get if you start a new reindex on master, then try to use it with 0.11.x or older
< wumpus> this is due to chainstate obfuscation
< dcousens> wumpus: indeed you are right :)
< GitHub197> [bitcoin] laanwj opened pull request #6933: build: Improve build instructions (master...2015_10_modernize_linux_build_instructions) https://github.com/bitcoin/bitcoin/pull/6933
< dcousens> wumpus: that monospace thread never stops haha
< wumpus> dcousens: yeah... it's surprisingly hard / impossible to get monospace font to work well across platforms, previous tries were eventually given up
< wumpus> making a cross-platform GUI is hard enough if you use only one, default font
< dcousens> well, default font is the hard part
< wumpus> nice, bitcoin-qt works as-is in QtWayland in Weston w/ no X server. Prepared for the future :)
< GitHub39> [bitcoin] jtimon closed pull request #6068: Policy: Create CPolicy interface and CStandardPolicy class implementing it (master...policy_new) https://github.com/bitcoin/bitcoin/pull/6068
< jonasschnelli> wumpus: Nice. Did you test wayland on ubuntu?
< wumpus> jonasschnelli: yep
< jonasschnelli> Need to do that as well...
< jonasschnelli> Qt is a good choice. I just made a hardware wallet frontend with Qt. All UX expectations could be satisfied (custom fonts, animations and transitions, etc.).
< wumpus> in ubuntu 15.10 you can just apt-get weston and you'll can lon into a (very limited) wayland env. It's also possible to have a gnome wayland session but wasn't able to get that to work
< wumpus> s/lon/log/
< wumpus> nice!
< morcos> wumpus: should i rebase my pulls that have merge conflicts due to reversion of medianTimePast. or is it relatively safe to assume that will be re-merged in the nearish future
< wumpus> morcos: it's safe to assume it will be rebased (but no one has put up a pull for that yet)
< wumpus> s/rebased/remerged/
< morcos> thanks
< GitHub128> [bitcoin] gmaxwell opened pull request #6934: Restores mempool only BIP113 enforces (master...undo_undo) https://github.com/bitcoin/bitcoin/pull/6934
< GitHub139> [bitcoin] SegfredB opened pull request #6935: 0.10 (master...0.10) https://github.com/bitcoin/bitcoin/pull/6935
< JoyStick> hey, any dev familiar with OS X version of 0.11.1 core?
< Luke-Jr> JoyStick: rule 1) never ask to ask, just ask; rule 2) if it isn't *development-related*, ask in #bitcoin
< JoyStick> asked. very few Mac users there. My issue is with the .dmg file which wont attach. Apparently there is checksum mismatch, but cant fix that one from this point. official support site is way too slow.
< Luke-Jr> JoyStick: lack of answers in #bitcoin is not a reason to ask support questions here.
< Luke-Jr> cfields: ping
< GitHub187> [bitcoin] morcos opened pull request #6936: [WIP] Keep pcoinsTip cache warm (master...HotCache) https://github.com/bitcoin/bitcoin/pull/6936
< dcousens> -given how confusing it can be to reason about, why would anyone use time over a relative block height in terms of [relative] lock time?
< dcousens> is it purely ot counter the variability of block discovery?
< phantomcircuit> sipa, how difficult would it be to do headers sync while we're reindexing and/or loading a bootstrap.dat ?
< phantomcircuit> (so we have the headers for the fScriptChecks flag)
< sipa> phantomcircuit: i've had a few ideas... one is to make reindex just not wipe the block index
< sipa> another idea is to make reindex just rescan for blocks and add them to the blockindex, but not validate
< phantomcircuit> sipa, and what just reset the validation states?
< sipa> no, wipe the chainstate
< sipa> the block index does not say what the best chain is
< phantomcircuit> oh right
< phantomcircuit> forgot about that
< sipa> only where which block is on disk, and how validated it is
< sipa> i like the rescan idea more
< sipa> it's more general
< phantomcircuit> sipa, yeah that's what i was saying, reset the chain state and the block validation state (the two separate things)
< dcousens> should a re-indexing node have a 0-size mempool?
< jcorgan> what, do we have github pull request spam now?
< phantomcircuit> jcorgan, yeah it's someone who doesn't understand the difference between a pull request and creating their own branch