log☇︎
1100+ entries in 0.087s
asciilifeform: mircea_popescu: trb is 100% sha currently
asciilifeform: wouldn't go this far; dunno about mircea_popescu , but i'm presently connected to fleanode, trb, etc via tcp
asciilifeform: would be neato imho if we could bring 2.6 back to life trb-style.
a111: Logged on 2017-01-03 20:33 mircea_popescu: otherwise it is discarded. B.T may be pruned (according to arbitrary address list, for instance). Rate limiting in TRB.N may be constructed to observe N.B items that fail to propagate to B.B and ban the originating peers.
mircea_popescu: http://btcbase.org/log/2017-01-03#1595770 << alluded to in the more specified TRB.B/TRB.N prototype design. ☝︎
mircea_popescu: http://btcbase.org/log-search?q=from%3Amircea+trb-i << interesting list, at that.
asciilifeform: iirc was one of the moar recent 'trb-i' mircea_popescu threads
mircea_popescu: but yes, this is the other half -- need to bake trb nodes that won't get insta-banned by wot-trb on sight because they spew garbage
mircea_popescu: http://btcbase.org/log/2018-09-24#1853601 << this seems so to me. how about we wot the trb, rather than iptables it. ☝︎
asciilifeform: sanely-behaving nodes (both trb and hypothetical trb-compatible heathens, which exist, or we'd see 0 blox) ought to be able to find ~and keep~ each other company
asciilifeform: imho a trb node ought to only advertise addrs that either a) part of the manual --addnode set the node was brought up with , or b) actually supplied a correct block in the recent past
a111: Logged on 2018-09-23 21:49 asciilifeform: in mostly unrelated lulz, there are apparently noads who shove a couple 100 MB (!) of bastard blox into a connected trb, prior to the latter throwing the ban switch ( because of the idjit shitoshi networking routine, actual disconnect happens a good 10-20sec after ban )
asciilifeform: mod6: i was thinking of trinque's idea : suppose trb closed all open pipes if it finds that $configurable hours (e.g. 3) have passed without new blox
asciilifeform: trinque: the crapola-gluetrap steadystate in trb is what i was trying to kill with 'wires' experiment
a111: Logged on 2018-09-23 21:49 asciilifeform: in mostly unrelated lulz, there are apparently noads who shove a couple 100 MB (!) of bastard blox into a connected trb, prior to the latter throwing the ban switch ( because of the idjit shitoshi networking routine, actual disconnect happens a good 10-20sec after ban )
trinque: the behavior I've seen is that this particular box gets stuck, always immediately resumes receiving blocks when trb is restarted
asciilifeform: it's a network problem, not a trb one.
trinque: kill trb, restarts, resumes eating
asciilifeform: in mostly unrelated lulz, there are apparently noads who shove a couple 100 MB (!) of bastard blox into a connected trb, prior to the latter throwing the ban switch ( because of the idjit shitoshi networking routine, actual disconnect happens a good 10-20sec after ban ) ☟︎☟︎
a111: Logged on 2018-09-08 23:31 mod6: Hi all, I wanted to get some review, on this manifest file that would be introduced into trb. It would be placed inside the 'bitcoin' directory along the 'Makefile' and 'src' directories. It follows trinque's proposal here: http://trinque.org/2018/06/02/v-manifest-specification
mod6: http://btcbase.org/log/2018-09-08#1848783 << >> http://p.bvulpes.com/pastes/rjqUZ/?raw=true << Back on the 8th I posted this possible manifest.txt file for trb. It's now been zapped, but I made a new one with updated block heights for all. ☝︎
asciilifeform: i can do it with shell ( as quoted from trb experiment list earlier ) easily enuff. but there is also the ~reader~ side, who wants to verify continuity.
asciilifeform: which complete kit oughta be able to follow the continuity of e.g. trb to day1.
asciilifeform: i put 9000 hrs into reading trb, i do not have these 9000 to do it exactly same again nao.
asciilifeform: mircea_popescu: canhaz link to 'hey folx, phf's vtron out of beta, let's regrind trb etc nao' ?
a111: Logged on 2018-09-19 14:42 mod6: Before we embark on an entire regrind of the trb-vtree to use keccac, I think we just need a major release version of a "defacto" vtron that supports both SHA512 (for other legacy projects) and keccac. Sounds like phf's might fit the bill, but want to ensure that when the Foundation tackles this problem, it's on very stable footing.
mod6: Before we embark on an entire regrind of the trb-vtree to use keccac, I think we just need a major release version of a "defacto" vtron that supports both SHA512 (for other legacy projects) and keccac. Sounds like phf's might fit the bill, but want to ensure that when the Foundation tackles this problem, it's on very stable footing. ☟︎
mod6: To me, I'd like to see just one trb-vtree, one set of patches. And if that means moving to keccac, then we will, and that'll be the only set of vpatches that The Foundation will distribute. (Otherwise I think it's too confusing, and time consuming to publish future vpatches in both.)
a111: Logged on 2018-09-19 13:04 asciilifeform: hey mod6 , didja ever switch to new format in trb ?
asciilifeform: hey mod6 , didja ever switch to new format in trb ? ☟︎
asciilifeform: btw one of the items i have in the deep freezers, is a trb with block db ripped out, replaced with (half-finished, sadly) ada mmap thing
asciilifeform: ( trb still has )
asciilifeform: i suspect the time to snapshot and pour cement, trb style, approaches.
mod6: http://p.bvulpes.com/pastes/qSPwN/?raw=true << draft of the manifest.txt for TRB. mircea_popescu asciilifeform ben_vulpes trinque et. al. plz to review and let me know if any thing needs changing.
asciilifeform: i can see an argument for keeping a cut-down version (perhaps trb's) around on ~some~ boxen , simply for curl-for-fetching-heathen-www, but that's about it.
asciilifeform: like it or not, 'faberge egg' is not a good weapon. to which i'll add that imho trb only recently (with 'aggression') graduated from 'faberge' status, previously noad required regular babysitting, and i've found that this is no longer so
mod6: A while back it was argued that a full-regrind of the entire trb vtree was not necessary. Having all the manifest entries in there, even if starting now, should suffice.
mod6: trinque: Thinking about it here quick, I think I ought to have all of them in there for what is included in a "proper trb". Adding them all in now, would give them all similar block times, but that's fine. It is what it is.
mod6: Hi all, I wanted to get some review, on this manifest file that would be introduced into trb. It would be placed inside the 'bitcoin' directory along the 'Makefile' and 'src' directories. It follows trinque's proposal here: http://trinque.org/2018/06/02/v-manifest-specification ☟︎
asciilifeform: observe that trb in fact worx pretty well, even though mircea_popescu did not lay cable.
mod6: afaik it's just running trb
a111: Logged on 2018-08-22 14:34 asciilifeform: in other heathen lulz, https://archive.is/yUvwS << 2017-style shitfork , rebranded as a kind of faux-trb, 'Satoshi Vision' of... 'block size to 128 MB'
asciilifeform: in other heathen lulz, https://archive.is/yUvwS << 2017-style shitfork , rebranded as a kind of faux-trb, 'Satoshi Vision' of... 'block size to 128 MB' ☟︎
asciilifeform: mod6: hey, at least wolf d00d didn't go to whatevercon in vegas and present trb as own phd work or whatnot
a111: Logged on 2018-08-17 15:08 mod6: ok sooo he did get a trb built from the guide and get it running. not sure why he never stopped in here. and ya, looks like he just grabbed everything and mirrored it on his site.
mod6: ok sooo he did get a trb built from the guide and get it running. not sure why he never stopped in here. and ya, looks like he just grabbed everything and mirrored it on his site. ☟︎
asciilifeform: mod6: http://edgecase.net/articles/2017-11-25_edgecase_datafeed_article_21_2017-11-25_stjohn_piano_compiling_bitcoind_trb_054_on_debian_711/rotor.tar.gz.asc , http://edgecase.net/articles/2017-11-25_edgecase_datafeed_article_21_2017-11-25_stjohn_piano_compiling_bitcoind_trb_054_on_debian_711/programmable-versionstring.vpatch , buncha others
asciilifeform: dredged up d00d via script that digs for refs to trb in heathendom. turns out, he hosts mirror of trb patches ( they dun seem to be discussed in articles tho, near as i can see )
deedbot: http://trinque.org/2018/08/16/new-trb-node-at-1610121250/ << trinque - New TRB Node at 161.0.121.250
asciilifeform: mircea_popescu: threading in trb is exactly variant of ye olde http://btcbase.org/log/2016-01-21#1379603 -- it's there because idjit had nfi how to do nonblocking i/o ☝︎
mircea_popescu: (yes, i'm using this to raise the army that'll surgerize trb, of course, of course)
asciilifeform: lol exactly like trb?!
a111: Logged on 2018-08-14 20:50 asciilifeform: http://logs.minigame.biz/2018-08-14.log.html#t01:52:40 << incidentally asciilifeform coupla yrs ago made failed attempt to replace 'boost' in trb with cpp11isms. principle obstacle was that gcc 4.9x doesn't really have 100% working cpp11.
asciilifeform: http://logs.minigame.biz/2018-08-14.log.html#t01:52:40 << incidentally asciilifeform coupla yrs ago made failed attempt to replace 'boost' in trb with cpp11isms. principle obstacle was that gcc 4.9x doesn't really have 100% working cpp11. ☟︎
asciilifeform: Mocky: cpp proggy always rides on libc. witness trb, the orig experiment with musl here.
asciilifeform: trb noad is up again as of nao.
asciilifeform: iirc Funkenstein remains the unbeaten recordsman there ( operated, at least at 1 point, a shitfork based on cribbed trb )
mod6: Alright all, the TRB vtree was updated with asciilifeform's ascii-armored pgp sig.
lobbesbot: phf: Sent 2 days, 0 hours, and 1 minute ago: <asciilifeform> plox to update stable trb tree in http://btcbase.org/patches from mod6's
asciilifeform looks at 'ircd-ratbox' , somewhat astonished at the mass, coupla MB of src (moar than, e.g., trb)
mod6: Well, as he said, I don't think he's interested in contributing to TMSR~ any longer. Even if he wants to follow trb, for instance, can do so from the BTC-DEV list.
asciilifeform: shinohai: can relax, nobody is out to eat you, and yer a+++ trb tester, my rating for instance stands.
shinohai: Well I stopped in here as it is the preferred forum for trb matters, but if my presence is not welcome - well then!
asciilifeform: !Q later tell phf plox to update stable trb tree in http://btcbase.org/patches from mod6's
mod6: Lords and Ladies of The Most Serene Republic: The Bitcoin Foundation's vtree has been updated with asciilifeform's Aggressive PushGetBlocks vpatch. The TRB-HowTo Guide has been updated as well.
deedbot: asciilifeform rated shinohai 3 at 2018/01/24 23:16:20 << heathendom newsdesk; pogotronics, trb, FG experimenter
asciilifeform: trb's footprint problems are 100% on account of heapism.
ben_vulpes: http://btcbase.org/log/2018-07-16#1834927 << i'd not distribute anything but trb patches to allcomers ~already~; if i made a useful thing i'd trivially share the source for it with l1 and rely on y'alls judgement as to whom to further share it with but i wouldn't concern myself with preventing leaks-to-kloinkers. beyond that, i share certain specific source with a subset of my own l1 and no further, with a ☝︎
asciilifeform: ( hence why asciilifeform did not say 'wtf, why you lot distribute binariola, why not post just src like trb ' )
asciilifeform: heathens are , as i currently understand, so unspeakably stupid that 'stealing fixed code' is not to date ever observed, they re-break as soon as they touch. ( any heathens stole trb for the shitfork warz, other than funkenstein ? )
asciilifeform: for all i know, somewhere there exists a d00d who tried to build e.g. trb on a vax. and thinks asciilifeform is idjit, because it dun go. how's that skin off my back.
a111: Logged on 2018-07-15 15:59 mircea_popescu: http://btcbase.org/log/2018-07-15#1834748 << trb node is one of the few items that ~can't~ be backed up in the usual sense. because what you get if you just blindly copy is a shitup not a backup.
a111: Logged on 2018-07-15 14:56 asciilifeform: and whole concept of 'backing up trb' strikes me as wrongheaded -- the most effective backup is simply a 2nd node.
mircea_popescu: http://btcbase.org/log/2018-07-15#1834748 << trb node is one of the few items that ~can't~ be backed up in the usual sense. because what you get if you just blindly copy is a shitup not a backup. ☝︎☟︎
asciilifeform: a new node plugged directly in ( i.e. over lan ) to a working trb node, with 'aggression' should sync in week or 2.
asciilifeform: and whole concept of 'backing up trb' strikes me as wrongheaded -- the most effective backup is simply a 2nd node. ☟︎
asciilifeform: shinohai: i've found that trb state can't be effectively backed up without stopping the node ( otherwise indices turn to barf . ) how do you do yours ? periodic stops ?
shinohai: SO have entire trb chain backed up in case of failure ( also www, etc)
deedbot: mircea_popescu rated dooglus 2 at 2016/03/07 09:36:24 << ran one of the best dice sites of all times, before being torpedoed by fiat government posturing. trb testing.
a111: Logged on 2018-07-14 13:24 brazilish: i'm trying to running trb on the cheapest possible setting
asciilifeform: http://btcbase.org/log/2018-07-14#1834457 << ftr this is a ~very~ Bad Idea . for instance , i do not know in what starvation-cheapo hoster brazilish put his node, but if it was aws he will find that he cannot connect to ~any of the l1 folx's trb nodes, most people ban aws ip range whole ☝︎
a111: Logged on 2018-07-14 13:37 brazilish: can i ask what specs have actual running trb nodes?
asciilifeform: http://btcbase.org/log/2018-07-14#1834483 << just about anything with 2GB or more memory will work -- trb is not really cpu-bound . a 1TB disk will last you 3-5 yrs until fills. ideally ssd ( though the exact penalty from using mechanical disk is not yet known, see last night's thread ). if ssd -- i recommend samsung's, cheapo ssd typically burns after 1-2 yrs of trb wear. ☝︎
diana_coman: uhm, search for "trb node" in the logs - that's about the closest you'll get
diana_coman: !!s trb node
brazilish: can i ask what specs have actual running trb nodes? ☟︎
deedbot: diana_coman rated brazilish 1 << He says he's trying to run TRB and get himself a blog.
brazilish: i'm trying to running trb on the cheapest possible setting ☟︎
lobbes: http://btcbase.org/log/2018-05-22#1816438 << in other news, you were not kidding! my trb (hdd + aggression) has moved a mere 20K blocks in one month ☝︎☟︎
mod6: I'm happy to announce that The Bitcoin Foundation's 2nd TRB node is fully sync'd and will be added to the list of Advertised Republican Nodeds.
asciilifeform: ( i vaguely suspect there exists even a 'herd immunity' of sorts, where aggressive trb helps other peers, of whatever config, stay in shape )
asciilifeform: afaik it doesn't auto-timestamp tho, iirc ben_vulpes , trinque , and possibly others, use linux system log to 'cheat' and get timestamps from trb
asciilifeform: i suggested it specifically to shinohai because he's apparently still experimenting with trb and afaik isn't tied up with anyffing else.
asciilifeform: ( point being, it may very well be feasible to run a reliable infrastructural trb noad on mech hdd. certainly worth test. )
asciilifeform: at any rate, the block processing delay was a kind of red herring, turned out that 'wait for rain to fall' was in fact the culprit in all ( known to asciilifeform ) cases of 'trb node won't stay in sync'
asciilifeform: notably , the patch started life as a general-purpose profiling of a whole buncha things in trb, but i ended up cutting out all but block timing, given as that was where something like 99% of ~active~ (i.e. as opposed to waiting for rain to fall, this was pre- aggression) cycles are spent
asciilifeform: diana_coman: reporting block r/w times currently requires pressing with the little patch that takes the times. ( it is not part of 'flagship' trb )
asciilifeform: i expect that soon n00b will be able to revv up a trb box simply by making use of trinque's builder and then 'emerge trb' or the like.
asciilifeform: rk3328-roc-cc + usb3sata snake + 1tb mech hdd -- might potentially make a working ~100bux trb node. worth a shot, shinohai , if you have a free hand.
asciilifeform: shinohai: prolly the best result re trb on rockchip would be with 'adult' samsung sata ssd via usb3-sata snake ( you can buy short one, that fits in box )