< luke-jr> huh, apparently there's a way to fsync just a single directory. I wonder if we should try to use that.
< luke-jr> $ xfs_io -c fdatasync /mnt/testdir
< luke-jr> (not limited to xfs)
< gmaxwell> I've unsubscribed from bitcoin-dev. Feel free to forward me anything the needs my attention.
< btcdrak> thank jeremyrubin
< GitHub154> [bitcoin] crowning- opened pull request #8903: Seed insecure_rand during start (master...patch-1) https://github.com/bitcoin/bitcoin/pull/8903
< SL__> what port is needed to be opened to run the BU node?
< achow101> SL__: probably 8333. This channel is for Bitcoin Core only, so discussion about other clients is OT
< SL__> Ok.. sry just clicked the link on BUs github. But thx anyway. :)
< GitHub156> [bitcoin] dagurval opened pull request #8904: [qa] Fix compact block shortids for a test case (master...shortid-coinbase) https://github.com/bitcoin/bitcoin/pull/8904
< dcousens> (assuming a transaction is to be encoded per BIP141), is witnesses length always == to inputs length? Just reading: "A non-witness program (defined hereinafter) txin MUST be associated with an empty witness field, represented by a 0x00."
< sipa> confused
< dcousens> sipa: indeed, I may be :P
< sipa> i really don't know what you're asking
< dcousens> wait, lets roll back, is the witness field prefixed with a varint to indicate length? That is probably a mistake that slipped in during a WIP on the code I'm looking at
< dcousens> (is the witnesses** field)
< sipa> for each txin, a txinwitness is serialized
< dcousens> sipa: to rephrase, [nVersion][marker][flag](varint)[txins](varint)[txouts](varint?)[witness][nLockTime]
< sipa> no, there is no varint before the witnessss
< dcousens> right, cheers, that explains the confusion :)
< sipa> it's implicitly the same count as txin count
< dcousens> yeah, thats what I figured
< sipa> ok