tmsr - Log 2017-10-20

mircea_popescu: and here's a bit of an evolutionary biology mindbomb : since the ability to distinguish one's own sons from other's sons is positively selected in males, it is altogether probable males have developed the ability to recognize their own spawn as such.

mircea_popescu: soo... does the new baby LOOK TO YOU like it's yours ?

shinohai: In other "Nobody could have predicted" http://archive.is/wTi4C

mircea_popescu: "here's a list of a half dozen tards who paid some scammers way too much money to be named in an inept piece in random fiat fake news outlet"

mircea_popescu: "*A substantial profit. Our ageing editor needed the phrase translating." outsourced writing, too.

BingoBoingo: In other lols, there is a place in Brazil that has datacenters and it is lulzily enough named Uberlandia

BingoBoingo: In other news Portuges feels more more Italian than Spanish on deeper digging

BingoBoingo: mircea_popescu: Enjoy not all the numbers. Apparently while life may be cheap in Asia, IPs are not? http://wotpaste.cascadianhacker.com/pastes/v35oI/?raw=true

phf: asciilifeform: don't forget to bring the customary earth in a small baggy from the motherland

shinohai: This is great ... what can possibly go wrong? https://github.com/Arachnid/solsha1

diana_coman: on Centos 6.8 (64bits) trying to read fg's output from ttyUSB0 with dd hangs on most occasions (erratically it will work but I haven't figured out when exactly); at the same time, a "screen /dev/ttyUSB0 115200 " works perfectly fine <-- anybody knows what this is about?

diana_coman: I might add that I *did* run the stty command first ofc; AND reported settings afterwards seemed fine

diana_coman: to re-phrase the question then: does anyone know of any centos/dd-specific idea re flow control? very same tty setup works on ubuntu or xubuntu verified; to reiterate: ttyUSB0 IS set with stty -F /dev/ttyUSB0 115200 raw -echo -echoe -echok; anything else centos wants there specifically?

diana_coman: otherwise I guess I'll have to figure out what screen does exactly that it works, ugh

mircea_popescu: iirc centos format is (sudo) setserial blabla, it lists devices / state etc with -g

diana_coman: hm, it sees it/reports it fine as far as I can tell; pl2303 attached to ttyUSB0; ehci_hcd

mircea_popescu: i would guess, since screen sees it fine, that your problem is later in the flow. maybe the windows manager fucks it up for you ?

diana_coman: will dig some more later I guess; unless someone knows this already

mircea_popescu: does problem persist if you use the actual system tty rather than X ?

shinohai: In CentOS I think one has to pass -ixon or -ixoff to turn flow control on/off

diana_coman: shinohai, that seems to be a bit closer but not yet solved: added -ixoff and next run of dd went fine but then it was stuck again; turned on and off again and it was still just hanging so simply reset on that was not enough either; something is weird there

shinohai: Indeed .... been a long time since I messed with CentOS so I may be wrong here ....

diana_coman: guess I'll go through the full list and at least see if there is anything that might help

diana_coman: eh, raw actually includes -ixoff and -ixany so no wonder that did not do much

asciilifeform: diana_coman: gotta find out how your setup differs from that of mod6 , trinque , ben_vulpes , danielpbarron , et al

asciilifeform: diana_coman: also double-check your wiring.

diana_coman: hence, more a matter of what does screen do differently

asciilifeform: what means screen works ?

asciilifeform: yes it sees the tty. but FG puts out ~arbitrary~ octets (yes all 8 bits) and the tty gotta work with this without losing what it thinks is 'control chars' (0x13, 0x03, etc)

asciilifeform: screen won't tell you this

diana_coman: asciilifeform, I'm not yet even there basically; dd simply hangs

asciilifeform: when you ctrl-c , what's it say ?

asciilifeform: '0 bytes copied' ?

asciilifeform: red lamp on FG lit ? flickers ?

diana_coman: one in x runs of dd will even actually copy something

diana_coman: green lamp steady on, red lamp flickering just like in *another* fg that is currently purring away nicely at another machine

asciilifeform: switch the serial cables from one to the other

asciilifeform: b/w the working and the sad one

asciilifeform: entirely possible that you got a defective cable, i do not test the cables, only the FG

asciilifeform: ( i expect that most purchasers throw away the usb cables )

diana_coman: did a stty -F /dev/ttyUSB0 -a on a working machine and on this machine, flags are same except for hupcl vs -hupcl on centos and onlcr vs -onlcr on centos; and a -cdtrdsr on centos, not mentioned on the other

asciilifeform: 1) switch the entire ensembles places

diana_coman: asciilifeform, 1. I tried already 3 other cables; unlikely all three borked

asciilifeform: 2) then switch back, and switch only the cables.

asciilifeform: screen will not necessarily show anything if you feed it 8bit garbage

diana_coman: but it shows...something at least; why does the other one not show anything?

asciilifeform: the canonical test is dd if=/dev/yourfg | hexdump -C -v

diana_coman: I switched this whole board to yet another machine , with its cable too; there it works

diana_coman: yes, canonical hence why I'm trying to get to the bottom of this

asciilifeform: at this point i'd say that ubuntu is not a linux and explicitly unsupported...

asciilifeform: it's a fuckknowswhat

diana_coman: asciilifeform, yes, fine; trouble is it's centos where the thing is not working

asciilifeform: hm now that's odd

asciilifeform: betcha it's a buggy pl2303 driver.

asciilifeform: do you have a ftdi or other type of ttlserial cable in the house ?

asciilifeform: i picked pl2303 because it is the most afaik widely supported, but apparently not even it worx everywhere ☟︎

diana_coman: hm, should have somewhere; will dig and try I guess

shinohai: Well how do you expect to get collared, slut, if you don't say anything?

BingoBoingo: Have your tried plugging it into your rs-232 port shinohai?

BingoBoingo: Your serial port, you know so she can speak through

phf: diana_coman: you can also try ``cu -l /dev/ttyUSB0 -s 115200'' since that doesn't take over the terminal AND if it works, you should be able to paste ``strace cu...'' output from which we can figure out what system calls that thing did before touching the device

asciilifeform: phf: i suspect this will do 0, in buggy pl2303 driver i encountered, it's the bandwidth and flowcontrol setttings that didn't take. but other than that adapter behaves normally

asciilifeform: which in the case of FG means worse-than-useless

asciilifeform: ( will pass through some bits, but at fraction of rated bandwidth and PATTERNED crapola )

asciilifeform: i've been thinking for entire year what is to be done about this, but have conceived of no solution other than THROW AWAY BROKEN OS OMFG

phf: well, if her stty is identical to whatever screen does, she should end up in an identical broken state, i.e. getting bits from dd same way she's getting bits from screen

asciilifeform: i don't see a mention of anything ~appearing~ in diana_coman's screen -- merely that it doesn't barf from nonexistent tty

asciilifeform: diana_coman did anything appear ?

phf: i assumed by http://btcbase.org/log/2017-10-20#1727227 she means that there's some kind of output ☝︎

a111: Logged on 2017-10-20 17:26 diana_coman: screen /dev/ttyUSB0 115200 reads nicely

diana_coman: asciilifeform, yes; I was trying to get that through but I'm possibly too tired for it: using screen results in constant stream of data

asciilifeform: pretty odd.

diana_coman: on the sad part it seems suitable cable not in house atm so it'll have to wait until tomorrow

phf: asciilifeform: you can have a hardware button fixed pattern: hold a button, a fixed pattern of bits come out that you use for sanity check

diana_coman: on the possibly enraging more digging turned out a known shit -difference between pl2303 and pl2303x ☟︎

diana_coman: apparently uncompatible though very similar and reported same ofc etc

asciilifeform: phf: on one hand could be useful, on other -- one more piece for enemy to wedge shut or otherwise manipulate in the post

asciilifeform: for enemy, for dust, for workings of ordinary mechanical failure

asciilifeform: i dun like having rng that is physically capable of putting out nonrandom.

asciilifeform: bad enough that the cpld fw is rewritable.

diana_coman: apparently there are 2 adapters namely ps2303 and ps2303x; incompatible; same vendor id and product id though

asciilifeform: ( i'dve installed onetimefused cpld, but IT DOES NOT EXIST on the market )

asciilifeform: nuts.

diana_coman: "the pl2303x can be distinguished from a pl2303 by checking bMaxPacketSize0 for device using lsusb -v -d 067b:2303; if bmaxpacketsize0 is 64 you probably have pl2303x"

phf: a programmer had a serial problem, "i know, i'll use usb to serial", now he had two problems ☟︎

asciilifeform: phf: i fucking hate usbtoserial dongles

asciilifeform: problem is that ~nobody has an actual serial port

phf: they are basically like flash drives, 50 different branded versions of the same 5 Chinese controllers, 3 of which report same signature but behave slightly different, the other two are bugged by the agencies

asciilifeform: afaik there is not a clean solution to this, other than to burn the pc nonsense to the ground

asciilifeform: i will not make rngs that DEMAND 2way protocol and usb micro ( with the rewritable firmware nonsense ACTIVE AT ALL TIMES , and 50MB pdf of liquishit spec with holes )

asciilifeform: and i will NOT make a rng that doesn't work with msdos box

asciilifeform: so it is and will forever be a serial device, compatible with e.g. data diode.

asciilifeform: who has a box that understands plain old circa 1960s rs232 -- can use fuckgoats

asciilifeform: who does not have -- cannot. ☟︎

asciilifeform: it is pretty basic 'pons asinorum' of sane hardware.

diana_coman: asciilifeform, I might be wrong on this but my current understanding is that it's my machine that has the *older* rather than the "required" newer driver

BingoBoingo: In other lol: "We don´t work with colocation solutions, it´s possible we work with hosting servers?"

asciilifeform: also i'm open to shipping something other than pl2303, if anyone has suggestions. ftdi is BANNED because 1) it costs 10x moar than pl2303 2) it is american/tw 3) it has rewritable fw that cannot be disabled

asciilifeform: BingoBoingo: 'sorry this is a tobacconist, not a colo' ? ☟︎

phf: прачечная хуячечная это министерство культуры

a111: Logged on 2017-10-20 18:52 phf: a programmer had a serial problem, "i know, i'll use usb to serial", now he had two problems

mircea_popescu: http://btcbase.org/log/2017-10-20#1727261 << where "everywhere" is one of the most widely installed web distros, at that. not even fucking obscure or anything. ☝︎

a111: Logged on 2017-10-20 17:44 asciilifeform: i picked pl2303 because it is the most afaik widely supported, but apparently not even it worx everywhere

a111: Logged on 2017-10-20 18:03 deedbot: collarmeslut voiced for 30 minutes.

a111: Logged on 2017-10-20 18:49 diana_coman: on the possibly enraging more digging turned out a known shit -difference between pl2303 and pl2303x

mircea_popescu: cuz obviously no standard can stay unembraced.

a111: Logged on 2017-10-20 18:56 asciilifeform: who does not have -- cannot.

mircea_popescu: http://btcbase.org/log/2017-10-20#1727313 << you'd be fucking surprised how often this is the case. "hello sir, this is bank, it just doesn't do any banking, problem ?" ☝︎☟︎

a111: Logged on 2017-10-20 19:01 asciilifeform: BingoBoingo: 'sorry this is a tobacconist, not a colo' ?

asciilifeform: again if mircea_popescu , diana_coman , or anybody else here, wants to suggest a replacement dongle -- i'm all ears. or better yet if we didn't even ship a dongle, every machine owner has responsibility to find some way to have a working serial port, it is not properly a problem in the scope of FG

mircea_popescu: local fucktards for instance explained to me "they can't receive wires from abroad." i'm like... so got any rostipollo then ?

mircea_popescu: asciilifeform simmer down, what're you so irritable foar.

asciilifeform: mircea_popescu: the dongle thing is 99++% of the fg support mail to date

asciilifeform: and i have nfi what do.

BingoBoingo: Reffer questions to the dongler? It's his profession?

asciilifeform: BingoBoingo: this re diana_coman thread

mircea_popescu: i don't expect there's much anyone can do. so far, we gather dataz.

shinohai: Hola sexylucy .... imagino que esta ausente por el momento, puedes esperar un momento?

jhvh1: shinohai: Bitstamp BTCUSD last: 5970.0, vol: 14654.47240142 | Bitfinex BTCUSD last: 5972.8, vol: 60631.91066838 | BTCChina BTCUSD last: 4251.858, vol: 0.00000000 | Kraken BTCUSD last: 5978.0, vol: 4411.05321761 | Volume-weighted last average: 5972.57295322

a111: Logged on 2017-10-13 01:54 mircea_popescu: trinque don't sweat it, she'll do it when i tell her to, what. tomorrow, monday, whatever.

mircea_popescu: !!withdraw 0.02 1ECQ9KkhYLyaHEbB9WHqbTiUmY5w69y6S2

mircea_popescu: the future of the human race emerges ever clearer.

mircea_popescu: if i dun have it i can't alter the article so they can't have to make wallets etc.\

trinque: I'm over here eagerly awaiting a printout, first non-hand-keyed withdrawal

shinohai would be thrilled if he figured out a viable business involving deedbot/tips

asciilifeform: dice!!111

mircea_popescu: shinohai the good news being that the juggernaught of the republic doth daily advance, so eventually it'll come.

mircea_popescu: ben_vulpes dja know the romanian joke with the bloke putting in just the tip ?

mircea_popescu: hawt girly at that difficult age, kinda would like kinda dun wanna. guy proposes he puts in just the head.

mircea_popescu: she eventually agress, and he fucks the living daylights out of her.

mircea_popescu: after they come to, she's like "hey, i thought you were going to put in just the head ?"

BingoBoingo: Eh, figured someone would want surprise, but $3840/month 8kva and 2x10bps links

BingoBoingo: Roughly the same range, and Yeah, I don't see reason not too. Should have a bunch more numbers early next week.

a111: Logged on 2017-10-18 18:52 mircea_popescu: ben_vulpes dja want an advance ?

BingoBoingo: Yeah, one would be helpful. Still probing for other costs that might try to do the surprise sex thing on my butt. ☟︎

BingoBoingo: This seems to be a good time for any Lords or L2's interested in homing a box to reach out.