Show Idle (>14 d.) Chans

← 2021-05-08 | 2021-05-10 →
billymg: good morning, asciilifeform. my crawler seems to have hit a spam vein on the network, total unique IPs in the db exploded over the last two days to over 100k (these all come only from what a node returns in a 'getaddr' request). of those, when subsequently interrogated, only 8580 respond with a valid version message
billymg: have a look at some of the results -- there's an interesting pattern wrt number of peers returned
billymg: the ~8500 "actual" nodes number seems to be inline with what heathen trackers report as the total number of nodes on the network (~9k), so i suspect the crawler is nearly complete in mapping out reachable nodes
asciilifeform: billymg: really only makes sense to add ip to db once it answers at least once (rather than immediately when reported as peer)
billymg: asciilifeform: sure, trivial to do that, but that wasn't my point
billymg: what i found interesting is that only a handful of nodes on the network actually return peers
billymg: and that of those, the ones that are not TRB, return a crapload of spam
asciilifeform: billymg: unsurprise imho
billymg: as if majority of nodes on network are in a sort of "client only" mode, receiving peers from this small group of mega spam nodes
billymg: asciilifeform: agree
asciilifeform: billymg: these are known as pseudonodes.
asciilifeform: aka sybils.
asciilifeform: billymg: btw 'Bad header length' is normally from timeout mid-reply
billymg: asciilifeform: good to know
billymg: asciilifeform: btw, reason i filtered those results only on peers > 1 is because all but handful of nodes returning only a single peer returned a non-self ip
billymg: not sure if this something you've observed before, where node returns only one peer in response to getaddrs, and that peer is self
asciilifeform: billymg: not observed this, but would guess some variety of pseudonoad
asciilifeform: (neither trb nor any variety of prb iirc behaves this way)
billymg: in my results 2691 nodes returning a single (self) peer
asciilifeform: billymg: did these have anyffin else in common ? ( useragent ? )
billymg: lemme check..
billymg: asciilifeform: nothing that's really jumping out, user agents seem inline with overall network distribution (i.e. mostly satoshi 0.20, 0.21, a few 19s, 18s, 17s)
asciilifeform currently has no solid theory for these
billymg: regarding the lack of surprise to the network being, in practice, centralized, this is true for folks following along, but i doubt it is widely known
snsabot: Logged on 2021-05-09 14:34:40 asciilifeform: billymg: unsurprise imho
snsabot: (trilema) 2015-07-08 asciilifeform: it's where i describe how bitcoin is not actually decentralized at all, but depends on a thoroughly usgificated network of political clocks
billymg: my goal in making this crawler is to get more "bitcoiners" running trb nodes, and i suspect some al gore / nate silver stats and infograffix will widen the pool of those who see there is a problem
billymg: and for me, just seeing that table sorted by len(peers) desc made it plainly, undeniably obvious that trb nodes are the only real nodes on the network
asciilifeform: billymg: the most interesting/sinister fact is that the miners aint exactly trb (given as they appear to enforce the segshit nonsense, so far w/out any sign of rebellion) but factually trb-compat. from a connectivity pov
asciilifeform: (or at the very least -- compatible w/ some client which in turn is able to speak to trb)
asciilifeform: 1 of asciilifeform's objectives in his (unfinished) attempt at billymg-style crawler, was to map out the net and infer whatever is possible to infer re block propagation and what sort of item the miners (or miner-adjacent publicly-facing noades) are using
billymg: asciilifeform: at the moment i've started writing the flask frontend for this so i can get the live updated stats up somewhere public
billymg: as soon as that's up will publish genesis for the crawler portion
billymg: probably the www too, why not
billymg: asciilifeform: from there if you added some watchglass methods for getting blocks i could then incorporate those into the crawler (if that is what you meant by trying to analyze block propagation)
asciilifeform: billymg: to request a block, you need its hash, and the latter gotta come from somewhere
billymg: ahh
billymg: so you need an actual node?
asciilifeform: or rather, not necessarily in realtime.
billymg: asciilifeform: could something like this be used?
asciilifeform: billymg: used where? trb supports no such thing
billymg: i mean in watchglass, just to get the hash of a block so to request it
billymg: for network scanning purposes only
asciilifeform: billymg: recall that 'nth block' is defined via a rather complicated algo (executing which is large part of what a btc client does..)
asciilifeform: if you want to scan for conformance to The One Troo Chain, gotta get the block hashes from an actual working client which tracks same
asciilifeform: imho the 'cement' hashes would be 'good enuff' to distinguish block-storing nodes from cheap pseudos
billymg: asciilifeform: yeah, that was the intent, just see if the node will send over a block, any block
billymg: but i didn't think of the 'actually valid' in true chain problem
asciilifeform: billymg: in trb logs i regularly see 'martian' incoming attempted-block hashes
billymg: asciilifeform: as in attempting to send you bogus blocks or requesting bogus blocks from you?
asciilifeform: billymg: sending
asciilifeform: (requesting nonexistent blocks aint a signallable condition, it happens regularly when diff. heights or 'frayed end')
billymg: gotcha
mats: whats the name of the phenomenon where folks are led to underprice risk because of safety features that reduce the need for operator attention (like tsla-style autopilot)?
mats: i remember a discussion from many years ago about the subject in the logs but i'm having a hard time recalling relevant search keywords
asciilifeform: wb mats . 'risk homeostasis'.
mats: thx pal
asciilifeform: mats: np
fghj: some aeons ago I tried to sync trb using 0.15 or 0.14 prb and it didn't work but 0.8.7 version could communicate with both prb and trb from what I recall
snsabot: Logged on 2021-05-09 15:05:08 asciilifeform: (or at the very least -- compatible w/ some client which in turn is able to speak to trb)
asciilifeform: welcome to #a, fghj . please briefly introduce yourself? (how found this ch?)
fghj: I'm long time lurker from bitcoin assets days
asciilifeform: fghj: indeed similar to my experience, the 10.x+ prb's tend to trip malleus and get disconnected.
asciilifeform: fghj: aite. welcome.
asciilifeform: fghj: btw there is also a (largely idle) #therealbitcoin .
asciilifeform: ( log , ftr. )
fghj: I think they stuff blocks with segshit data
asciilifeform: fghj: segshit produces blox that are compatible (in the non-forking sense) w/ trb. simply shows up as 'anyone can spend' tx.
asciilifeform: (also produces auxiliary turds that aint edible by trb, sent separately. hence 'segregated')
fghj: I coudn't even manually feed blocks from prb to trb using bash one liners and rpc calls
asciilifeform: fghj: iirc the liquishit gets stored past the 1e6byte mark. so theoretically could transform.
fghj: cool will maybe try in future
asciilifeform: i can't think of any reason to hand-feed prb-digested blocks into trb, tho
fghj: I was on slow connection and using spinning rust
asciilifeform: fghj: fwiw i still have a torrent up, with (most of, to date) blocks in ready-to-eat db
snsabot: Logged on 2021-02-25 16:29:31 asciilifeform: dpb, shinohai, et al : and lemme know whether worx. plox to seed.
asciilifeform: ( a number of people asked for it, given the atrociously slow default sync mechanism, and i finally caved & baked one )
fghj: ok, gotta get some zs now, will at some point erect server again
verisimilitude: It's a joke to call OpenBSD well documented.
verisimilitude: Using two keyboard layouts across two keyboards may as well be unsupported, and the man pages are useless.
trinque: !w poll
watchglass: Polling 16 nodes...
watchglass: : Could not connect!
watchglass: : Could not connect!
watchglass: : Could not connect!
watchglass: : ( Alive: (0.082s) V=99999 (/ Jumpers=0x1 (TRB-Compat.) Blocks=682824
watchglass: : Alive: (0.142s) V=99999 (/ Jumpers=0x1 (TRB-Compat.) Return Addr= Blocks=682674
watchglass: : ( Alive: (0.083s) V=70001 (/ Jumpers=0x1 (TRB-Compat.) Blocks=682824
watchglass: : ( Alive: (0.113s) V=99999 (/ Jumpers=0x1 (TRB-Compat.) Blocks=682824
watchglass: : ( Alive: (0.160s) V=99999 (/ Jumpers=0x1 (TRB-Compat.) Blocks=682824 (Operator: asciilifeform)
watchglass: : Alive: (0.178s) V=70001 (/ Jumpers=0x1 (TRB-Compat.) Blocks=682824
watchglass: : Alive: (0.083s) V=99999 (/ Jumpers=0x1 (TRB-Compat.) Return Addr= Blocks=682824 (Operator: whaack)
watchglass: : Alive: (0.166s) V=99999 (/ Jumpers=0x1 (TRB-Compat.) Blocks=682824
watchglass: : Violated BTC Protocol: Bad header length! (Operator: trinque)
trinque: huh, that's weird
trinque: the thing's up on my end, eating blox
whaack: asciilifeform: I've been meaning to ask, what does the Return Addr= and Jumpers=0x1 mean?
whaack: also, does anyone have a guide handy for setting up sbcl on one of ascii's gentoo distros?
trinque: no guide, but sbcl can bootstrap from clisp
trinque: such that you don't have to use some mystery meat binary to do so.
billymg: whaack: jumpers advertise "services" of a node, see
billymg: and per watchglass code, return addr is the responding node's 'addr_from' (the IP it claims as its own in its version message), which watchglass displays only if different than the ip used to connect to that node
whaack: billymg: Ah thanks, so then I guess the return address may be because I'm running on the same network as watchglass
billymg: whaack: strangely enough i'm seeing the same results for your node on my end
gregorynyssa: << never heard of that comic until now. this instalment was genuinely funny:
snsabot: Logged on 2021-05-09 14:35:43 asciilifeform: ( see also illustration. )
gregorynyssa: (unfortunately, most of the other instalments were too gross and lascivious to read.)
gregorynyssa: << is the term pseudo-node synonymous with "light node," or does it refer to a class of programs created specifically for nefarious purposes?
snsabot: Logged on 2021-05-09 14:35:15 asciilifeform: billymg: these are known as pseudonodes.
trinque: gregorynyssa: oh stop
trinque: "gross and lascivious"
snsabot: Logged on 2021-05-09 21:32:51 trinque: "gross and lascivious"
trinque: is there a religious reason for the aversion? I find most of her comics downright hilarious.
trinque: << possibly one of the best
gregorynyssa: << hard to say. I feel as if, I did not so much conform to a religion, as notice the existence of one or more religions which conformed to me.
snsabot: Logged on 2021-05-09 22:52:17 trinque: is there a religious reason for the aversion? I find most of her comics downright hilarious.
trinque: this was discomfort at other people's fun, or what?
gregorynyssa: << not really. are you deliberately trying to reference that quote by HL Mencken?
snsabot: Logged on 2021-05-09 23:07:21 trinque: this was discomfort at other people's fun, or what?
trinque: I'm unaware of the reference.
trinque: ha, but that's well said.
trinque: I imagine puritanism will be resurging in popularity after this meager plague.
← 2021-05-08 | 2021-05-10 →