Show Idle (>14 d.) Chans

← 2019-01-13 | 2019-01-15 →
asciilifeform: mircea_popescu: this is a mega-point
asciilifeform: aand i'ma apply the supplied algo.
shinohai: <<< Thanx, also will report to diana_coman if configuring out bullet has any negative consequences, so far everything renders correctly as far as I can tell.
a111: Logged on 2019-01-13 20:50 mircea_popescu: nice going shinohai
shinohai: Perforemed bare hand exploration, also works.
diana_coman: shinohai, an Ada lib for client-side logging would be quite useful
diana_coman: bot is in dire need of proper logging and once the new protocol is in place I can see the client itself needing to log stuff
diana_coman: at the very least, dubious messages if/when it gets them
shinohai: I look forward to client updates, also found quite interesting btw.
diana_coman: thanks! one of those days I'll get around to write-up the jam stuff for linking the lib with the client too ; at any rate, the linking seems sorted for now - it's Ada code for all the pieces that is missing
shinohai: While I'm here, did you guise remove permanently? (Perhaps jurov has copy still hidden away somewhere, I was interested for archival purposes)
mircea_popescu: should be in i don't expect anyone deleted it
shinohai: I tried cloning, simply times out ( was going to add additional post on subject )
diana_coman: wasn't that with old versions/unmaintained ?
mircea_popescu: diana_coman wth happened there, Trying Connection timed out etc. site got nuked ?
diana_coman: mircea_popescu, neah, it's that part that I did not move over to the html server because it wasn't up to date really
mircea_popescu: but i mean no part of the site responds ?
mircea_popescu: oh, is this uy1 being down ?
diana_coman: ? it responds here?
diana_coman: or uhm, no
diana_coman: yes, it is uy1 being down
BingoBoingo: mircea_popescu: .247 is uy1
mircea_popescu: asciilifeform BingoBoingo yo, you can't have multi hour outages, wtf.
BingoBoingo: asciilifeform: How is the debugging and instrumenting going? Do you need my hands in the rack?
shinohai: ( I wrote initial post on Friday, wasn't responding then when I did post research - only thing I got was the logs subdomain)
mircea_popescu: shinohai multiple layers. it went from .bz to .biz sometime 2017 maybe ?
mircea_popescu: jurov used to maintain a git repo (2016 maybe ?) but we're not that git-friendly anyway and he kinda stopped working on it.
asciilifeform: BingoBoingo: yes hands asap plox
mircea_popescu: when diana_coman moved over the site to uy, she pruned all the shit that was badly out of date and little hope of being brought back up
BingoBoingo: asciilifeform: Aite
asciilifeform: mircea_popescu et al : not a planned outage, at this point i strongly suspect iron issue
diana_coman: basically we don't have a windows maintainer anymore; and yes, last parts in there were ~2016
mircea_popescu: that's nice and all, but meanwhile here i sit talking to people about absent sites yes.
mircea_popescu: << in utter lulz : could the democrat party be any less fucking relevant these days ? it's gop-senate vs gop-house and trump.
mircea_popescu: bernie sanders is actually more of a player than the entire collected dnc
mircea_popescu: ie, pretty much ended as a going concern.
asciilifeform: BingoBoingo: wtf, thing appears to be hung on the grub menu screen
asciilifeform: update : it aint hung, responds, but i have nfi how this thing ever auto-booted in the past -- the boot menu dun seem to have a timeout , wtf
BingoBoingo: asciilifeform: In the rack
BingoBoingo: Please deliver instructions in #pizarro as I don't know how much log troubleshooting is going to generate
mircea_popescu: well, if you feed it kbd it will turn auto-timer off.
asciilifeform: mircea_popescu: hrm, i switch this off in my installs, but this box is a BingoBoingo creation.
asciilifeform: anyways booted nao, about to instrument.
asciilifeform: update : box is up; loox like BingoBoingo hastily reassembled thing last night without the nic plug properly seated.
mircea_popescu: you can do that ?!
BingoBoingo: I forgot to plug the ethernet back in
asciilifeform: haste makes waste, lol
asciilifeform: i'ma cure the asinine grub keyboard thing.
asciilifeform: ( who the FUCK thought this was acceptable behaviour )
mircea_popescu: desktop use.
asciilifeform: yea cuz it's a-ok to have desktop spin 4evah on boot... lol
mircea_popescu: making linux accessible to senior citizen drivers, whadda ya want
mod6: mircea_popescu: I've only seen behaviour from trb like that when I've had some bad hardware. In particular a suspected bad SSD. I issue the 'getinfo' command and nothing, just hangs.
BingoBoingo: << They just don't want anyone booting into the wrong version of Windows
a111: Logged on 2019-01-14 15:34 asciilifeform: yea cuz it's a-ok to have desktop spin 4evah on boot... lol
mircea_popescu: mod6 yeah but the disk is fine.
mod6: My one node I've got now, is still on that bad hardware as I haven't found a new home for it yet. Sometimes, if I just let it "figure itself out", it'll return eventually and be fine. I agree though, would be nice to get to a root cause of this by some form of testing. Hard to mimic the state, regardless.
mod6: And by 'eventually', I mean hours usually.
mircea_popescu: this had ~two weeks.
mod6: Oh, ok. Hmm. Well, thanks for reporting. We need to get to the bottom of that pile of weird.
mircea_popescu: and bad ssd is unlikely to hit ~specifically one~ of multiple instances, as well as it's unlikely to occur while drive has free spare sectors to map to, and besides it'd be a first. drive's fine what.
mod6: Alright, noted.
asciilifeform: update : uy1 is up, i have realtime log going from it.
asciilifeform: inspection of logs on disk turned up nuffin enlightening.
diana_coman: funnily enough it seems I even DID port over jurov's page too anyway, lolz; it was just the uy1 down thing
asciilifeform: update: aaand ben_vulpes (who is orig builder of UY1) didn't see it fit to build the k10temp kernel mod, so i can get cpu/chipset temps and ps voltages.
diana_coman: just not the git
asciilifeform: i'ma have to rebuild kernel on UY1 today.
asciilifeform: will announce the reboot hour in advance.
asciilifeform: anyone have objection ?
diana_coman: fine with me
asciilifeform: after we get to the bottom of UY1 issue, i'ma make sure that all iron owned by pizarro has asciilifeform-baked kernel in place.
mod6: asciilifeform: good with me.
asciilifeform: ty mod6 diana_coman
asciilifeform: BingoBoingo et al : also i am getting multi-second ping latencies to pizarro (all boxen)
asciilifeform: the debug session has distinct 'mars lander' feel to it
BingoBoingo: asciilifeform: I am seeing much faster pings
mod6: pings to from me are: round-trip min/avg/max/std-dev = 194.551/195.055/198.219/0.723 ms - 0 packets lost.
asciilifeform: mod6: 0.7s is pretty martian
asciilifeform: and quite atypical.
asciilifeform: well, moar moon than mars. but still.
asciilifeform: i am seeing multi-second ssh keystroke delays.
mod6: I don't doubt. That seems off for sure.
BingoBoingo: I'm seeing sub 20ms pings to every box I've tried so far except UY1. UY1 is returned sub 40 ms pings
BingoBoingo: From habitation module, not the rack
asciilifeform: BingoBoingo: soon as this is done, i'ma graph the machine temps/voltages.
asciilifeform: BingoBoingo: plox to keep kvm hoses in until further notice.
BingoBoingo: asciilifeform: ty, will keep the hoses in until you give the withdrawal order.
asciilifeform: ACHTUNG UY1 users : plz prepare for scheduled reset in 30m.
asciilifeform: incidentally i was not aware that UY1 uses grub (grub2 no less). grub is fatwa'd on pizarro-owned iron. i'ma correct this omission in a scheduled maintenance after we cure the moar pressing issue in re uy1.
asciilifeform: attn: rebooting UY1.
mod6: ty for the notice
asciilifeform: update : sensor reader proggy nao installed on UY1, will set to snapshot erry 30s.
BingoBoingo can see UY1 hosted sites
asciilifeform: so far all sensor readings within range
asciilifeform: and no peculiarities in log
asciilifeform: all fans moving.
asciilifeform: cpu temp 45.5C (out of 85 permitted), chipset 30.8 (75 permissible) (i'ma omit further reports on subj for nao, unless thing actually is found to melt)
feedbot: << PizarroISP -- Pizarro ISP Update January 14, 2019
asciilifeform: achtung pizarro subscribers : if 1 of you is running saturation test on the pipe, plz nao & in future ~announce prior~, ty.
mircea_popescu: very unlikely anyone's doing such a thing. you're prolly being attacked or w/e.
asciilifeform: mircea_popescu: it is entirely possible, i have not yet looked in detail.
asciilifeform: just nao wrapped up setting up realtime voltages/temperatures/etc output.
asciilifeform: the req still stands tho.
asciilifeform: mircea_popescu: if ddos, then nao as good time as any, it hasn't afaik yet been tested, how said pipe behaves under ddos.
asciilifeform: so far i've uncovered 0 that would reveal why box reset. but i expect before nightfall to find out, if iron is indeed misbehaving.
mircea_popescu: what's the book on that ?
mircea_popescu: i'd like to lay a bet :D
asciilifeform: i hate to disappoint mircea_popescu , but i'm all outta sheep for the haruspexy room, so cannot give odds lol
asciilifeform: now, elsewhere in conveyor :
asciilifeform: << this is not only troo but is how one tests a m-r ( you feed it known liars & known troofers for a particular N and verify output )
a111: Logged on 2019-01-13 21:13 mircea_popescu: << with some caveats as to how "unrelated" (in the m-r sense of the term) those candidates must be, as it's entirely possible to generate an infinite set of lying witnesses for any composite number.
asciilifeform: which is why mine reads witness as param rather than directly from rng
asciilifeform: at all times operator must be able to verify that m-r behaves as m-r.
asciilifeform: and indeed the quality of rng is 'jesus bolt' when running m-r in battlefield.
asciilifeform: ( recall, diana_coman uncovered various lulz re koch's variant )
asciilifeform: i'ma detail this subj in the customary pedantic detail in ch16 article.
asciilifeform: ( readers should also prepare themselves for following proof of why m-r worx, i'ma not simply cite the lit )
mircea_popescu: perfect.
asciilifeform: re verifications, observe the general pattern in ffaism: i expect any serious user to have a private battery of tests to verify that his, particular, ffa builds, actually conform to the declared behaviours.
asciilifeform: incl. the rng ( this is why the thing takes an arbitrary unix path for rng dev )
asciilifeform: ( yet aaanother thing that koch didn't give )
asciilifeform: mircea_popescu: observe that you can trivially generate an arbitrary-long sequence of liars for any composite N.
asciilifeform: 1 interesting maffs problem , is whether you can generate a sequence of witnesses that will tell troof of compositivity for ~any~ externally supplied N , up to given bitness. ( and as i understand this q is isomorphic to bach's generalized-riemann item that diana_coman dug up. )
asciilifeform: we will not be making use of this approach tho ( it relies on conjecture . )
asciilifeform: iirc there's a sequence that nails all composites < 2^64 . so could be handy for primorial generators ( vs. eratosthenes etc ) .
asciilifeform: but again , not directly pertinent to ffalogy.
mircea_popescu: it is, we had an entire discussion then
asciilifeform: i meant bach et al.
asciilifeform: asciilifeform's recommended recipe for m-rism still remains 'feed actual rng for witnesses'
asciilifeform: ( for folx who haven't yet read diana_coman's series -- the theorem diana_coman turned up, turns m-r into a deterministic ('aks but fast') test, iff generalized-riemann )
asciilifeform: found the old mircea_popescu thread, btw :
a111: Logged on 2017-10-08 19:15 mircea_popescu: asciilifeform (guy named arnault gave example of number for which all tests up to ~300 were misleading)
asciilifeform: ohai Mocky
Mocky: hiya
asciilifeform: what's new, Mocky ?
Mocky: fingers crossed maybe I get a job offer today
mod6: best of luck to ya!
Mocky: thx mod6
asciilifeform: update re uy1 : 2hrs and no anomalous readings. ( will revisit subj later tonight )
asciilifeform: in other vintage lultidbits, found while digging logs,
a111: Logged on 2017-10-09 16:39 asciilifeform: ... it follows that a 0.85sec 4096b modexp is all you need for a reasonable 'rsa phone' item.
feedbot: << Qntra -- DNA Pioneer James Watson Stripped Of Chancellor Emeritus Title Over Afropessimism
BingoBoingo: Latest old man dekulakized
asciilifeform: BingoBoingo: ha, i had nfi he still had any epaulettes or buttons to rip off
BingoBoingo: Well he had the symbolic "any key" that wasn't connected to anything
mircea_popescu: "afropessimism" actually great coinage.
asciilifeform: aka afrorealism..
mircea_popescu: never anything else.
mircea_popescu: i dunno why they're so butthurt by the ~statistical~ fact. just because average black is going to be dumber than average human doesn't say any one black person's gonna be dumb or anything like that.
mircea_popescu: they're not very appreciably butthurt at all by how average whitey's gonna undertest average "azn" aka chinese.
asciilifeform: there also butthurt
asciilifeform: !#s qntra harvard
a111: Logged on 2018-06-18 15:51 deedbot: << Qntra - Harvard Exposed Systemically Discriminating Against Asian Applicants
asciilifeform: & elsewhere.
mircea_popescu: i dunno, i guess my unexamined impression is otherwise.
Mocky: they think in terms of nanny state policies and what policies statistical facts would imply
mircea_popescu: but hence my objection ? turning a blind eye to all the chowdrys taking over bureaucracy while crying bloody murder over ~explanation~ as to why first black president === worst president ever ?
mircea_popescu: something dun wash.
Mocky: if they are gonna look at afrofacts, what next, look at femfacts?
mircea_popescu: so somehow "you can't run for city council, you're not 2nd gen azn immigrant" not quite as threatening as "you're fat because you eat too much and never get off ass" ?
asciilifeform: the chowdhrys are, as i currently understand, the vaisya/sudra-heavy wave of emigres & their nth gen, largely content with 'middling' ranks, the native usg.brahmin spawn dun want these
mircea_popescu: so basically they imagine they're getting a buncha eunuchs outta the deal ?
asciilifeform: same way they dun want into the academitardia racket ( when asciilifeform grunted in the ranks of said racket, was often the 1 euro face in the group pics.. )
mircea_popescu: i dunno, at this juncture i'm starting to suspect attemtping rational analysis is eminently wrong approach, not like it's a rational process.
asciilifeform: i was aboutta say
asciilifeform: it aint clear to me that 'they think' is a valid hypothesis, period
asciilifeform: they , in words of al schwartz, 'hind gut ferment'. they metabolize, but no thinking is necessarily involved
mircea_popescu: set of fashionable howls.
mircea_popescu: cuz they're a buncha fash dorks, what.
BingoBoingo: <mircea_popescu> so basically they imagine they're getting a buncha eunuchs outta the deal ? << I suspect this is an element of it.
BingoBoingo: Then there's the trying to expensively defend cheap respect for "all"
BingoBoingo: In other updates Pizarro www now has selection js in place when viewing individual blog entries
asciilifeform: BingoBoingo: confirmed, worx, a++
BingoBoingo: ty for the confirmation
asciilifeform: in other olds, 'The 256-bit integer q = (2x + 1)(4x + 1) with x = 0x400286bac15132db85b1c936709f369b passes 15 rounds of GMP’s primality test mpz_is_probab_prime_p...' ( Albrecht et al, aug. 2018 , / )
asciilifeform: ^ spoiler: gmp uses prng for witness gen
asciilifeform: mircea_popescu: after ffa is fielded, i'ma add gmp etc carmichaelisms to phuctor.
asciilifeform: would be surprised if it doesn't yield a couple thou lulz.
mircea_popescu: not a bad idea.
asciilifeform: there's a 1024bit example in that piece, and coupla others
asciilifeform: in other noose, 6hrs of reactor^H^H^Huy1 gauge , and 0 anomalies
asciilifeform at this rate might end up adding a 7th lcd to the wall..
mircea_popescu: asciilifeform do you keep a kernel hash or anything that'd expose a case of ?
asciilifeform: mircea_popescu: indeed i do
asciilifeform: i can't picture that any literate folx don't
mircea_popescu: and it's negative ?
asciilifeform: to date.
mircea_popescu: so canary yes, k10temp no ?
asciilifeform: i like to hash e.g. raid card biosen, etc., also
asciilifeform: mircea_popescu: so far all i know is that the thing rebooted without sanction, no fewer than 3x.
asciilifeform: (in span of 2 day)
asciilifeform: BingoBoingo reports intact seals.
asciilifeform: even if i turn up nuffin in next day or 2, i'ma have at least 1 moar planned reset, see whether ben_vulpes remembered to enable eccism when he plugged the thing in
asciilifeform: for some reason it is usually switched off by default on a new bios battery
asciilifeform: ( why , i cannot begin to fathom )
asciilifeform: for comparison, the 1 time dulap-III rebooted without sanction, was in may, when BingoBoingo elbowed the mains cable
asciilifeform: all machines of same type (varying only in ram/disk config)
asciilifeform: relatedly, asciilifeform has a backburner affair where disasm of the particular 3ware bios , looking for lulz (to date found none, aside from it having a quite bulky cpp runtime lib)
asciilifeform: so far the old proverb 'на ловца и зверь бежит' dun seem to play out for asciilifeform , somehow hitler dun feel like putting the delish booby catalogue items where asciilifeform's dirty orc hands can get to'em..
BingoBoingo: <asciilifeform> mircea_popescu: so far all i know is that the thing rebooted without sanction, no fewer than 3x. << It breathes ever so slightly better than it did before the opening and fan check
asciilifeform: BingoBoingo: y'know , these things ~do~ like a ~yearly breath of high pressure freon
mircea_popescu: i dunno, man. if i ran a moron empire with delusions of "pen devices" etc, it'd do precisely the sort of inept rebooting.
asciilifeform: ( in torture room, asciilifeform has an antistatic vacuum thing and other nice things )
asciilifeform: mircea_popescu: i dun see wai not to bug upstream, but then again i dun work at reichskanzlei
mircea_popescu: "it's not a truck" or similar reasoning.
asciilifeform: i've mostly sworn off trying to picture the 'reasoning'
asciilifeform: there's ~0 reason to it.
asciilifeform: mircea_popescu: if we live to bake the fpga router thing, will be interesting to give it a ring buffer that'll hold coupla 100MB of frames, and dump'em to flash upon any unsanctioned reboot of attached irons.
asciilifeform: oughta trivially capture 'magic packet' (under the old hypothesis that these result in fandango)
asciilifeform: i'ma take the chance to highlight the thing -- on sane iron , there'd be no such thing as a 'oops rebooted but fuck if i tellya why'
a111: Logged on 2019-01-14 04:15 mircea_popescu: i'd like a computer where i didn't have this sort of problem.
asciilifeform: there's no physical or practical reason why the last million or two instructions executed prior to whatever fail, should not be available after
asciilifeform: ( on top of ram contents, etc )
asciilifeform: 1 of the 9000 sane things bolix tried, valiantly , to do, within the limitations of the period irons.
asciilifeform: ( had 'fep', a supervisor cpu thing, but not the 'remote booby' item presently extant in intel irons, but a thing available strictly from console, and 'crash' halted the main cpu and dropped you into it, where you could peek/poke, probe irons, etc )
asciilifeform: fep ran from rom and could not be crashed other than by hand (or when iron is finally smoking, smouldering)
asciilifeform: asciilifeform finds it fascinating that even satellites, moon buggies, etc. naodays dun have a 'snapshot state of iron if crash'
asciilifeform: evidently it cannot be bought with moneys anymoar
asciilifeform: apparently this is what you get if there aint a 'lavrentiy pavlovich' in the design loop.
asciilifeform: imho if satellite 'oops it crashed and i cant tellya why', designer oughta get a 1way ticket to its orbit to go and debug..
asciilifeform: meanwhile, in darkest lulzafrica, 'Today we’re happy to announce that Urbit’s address space, now called Azimuth, is live on the Ethereum blockchain... We want Urbit to be a reliable, durable, permanent computer that’s simple to use. Today Urbit is still a prototype, not a consumer product. But we’re making great progress'
shinohai: "Buy our vaporware today!"
asciilifeform: 'that is not dead which can eternal lie..' or how did it go.
asciilifeform: ohai shinohai
shinohai: Heya asciilifeform
asciilifeform: shinohai: nifty eulora piece. yer ahead of asciilifeform in that battlefield, i've yet to attempt building eulora
shinohai: I still have nfi why the bullet barf, diana_coman says it has to do with proximity things. Hopefully I'll get some time in game to see if negative effects occur.
shinohai: (Also have companion trb piece, but was waiting for mod6 's keccak regrind)
asciilifeform: shinohai: neato. what didja do re trb ?
shinohai: Haven't completed build yet since mod6 published, but have tried the old method which unsurprisingly went fantastic.
shinohai: Cuntoo is simply a joy to use.
mircea_popescu: /mode #trilema +b *!*
mircea_popescu: << those poor people, and what they had to do for a logarithm.
a111: Logged on 2019-01-14 00:36 feedbot: << Bimbo.Club -- Philosophical Transactions. For the months of April, May and June, 1714 - Part I.
hanbot: well, this is my 3rd iirc restart of znc in two days, slightly more interruption than the problem i started using znc on UY1 to fix posed. looks like no prognosis yet eh?
lobbesbot: hanbot: Sent 1 day, 10 hours, and 22 minutes ago: <billymg> mp-wp patches have been updated: -- the only \ No newline at end of file strings you should find in .svg refs patch are from before part of the diff (i.e. file diffed had no newline at the end)
asciilifeform: hanbot: when was 'this' plox ?
asciilifeform: box is still up
BingoBoingo: hanbot: The investigation is ongoing. Other than the chassis interor being marginally cleaner than before asciilifeform instrumented the machine, answers remain elusive. Per I am inclined to not charge any Pizarro shared hosting customers for the month of January though I am open to hearing other suggested remedies.
asciilifeform: been up continuously since i set up sensors earlier today, and still alive, with 0 anomalous readings
asciilifeform: hanbot: we dun know yet wtf reset the box ( and it happened no fewer than 3 times, in 2 day span )
asciilifeform: hanbot: today i set up realtime stream of system log + voltages + temperatures + fan rpm to the torture room, was expecting to find thermal problem, so far 0
asciilifeform: hanbot & other subscribers to uy1 : plox to inform asciilifeform asap if you notice ~any~ unusual behaviour on this box ( not only reset , reset will be obvious from here )
trinque: conspicuous bit is various folks having moved their comms aboard uy1
BingoBoingo: During palm touch tests before cleaning fans the warmest part of the chasis was near the RAID card, by a margin that though small registered on my skin. Do we have a way to instrument the RAID card.
asciilifeform: BingoBoingo: it's in the log
trinque: but you know, pattern-seeking primate amidst randomness
trinque: ftr I don't need a month's comp for a few hours of outage, though a few hours of outage does suck.
asciilifeform: outages suck, period
asciilifeform: !#s how many bugs tolerate
asciilifeform: my fiber has 5-10 minutes of outage in a year, and ~that~ palpably suxx
mircea_popescu: the one concerning bit is whether indeed pizarro still owns that box or not.
BingoBoingo: trinque: My though on the month is that the money being paid for shared hosting is very real to our customers, and we lack a firm hour count on how many customer uptime hours have been lost.
asciilifeform: mircea_popescu: this q can be asked re any box.
BingoBoingo: <mircea_popescu> the one concerning bit is whether indeed pizarro still owns that box or not. << This very much concerns me
mircea_popescu: asciilifeform well no, original box had a chain of custody, and a history of predictable behaviour.
asciilifeform: mircea_popescu: fwiw crashism is more typical result of failed diddling than working.
trinque: BingoBoingo: signaling my willingness to pay, not my condoning of the outage.
mircea_popescu: this stance is consistent with, inter alia, republican practice -- we moved variously boxes off providers who kept rebooting them "mysteriously"
asciilifeform: mircea_popescu: i can't disagree, and am inclined to move it to cold spare when we get another crate in.
mircea_popescu: and the coincidence is there, and glaringly. we know for a fapt inept efforts ~signature move~ is rebootage, much like olde smersh signature move was the clicking on phone line ; and as he points out, buncha people moved their zncs there.
asciilifeform: the sad part tho is that if asciilifeform threw out erry retardix box that ever kernel panicked, would have none left in service
mircea_popescu: is this a fact, eg the rockchips ?
asciilifeform: i haven't succeeded in crashing a rockchip yet ( outside from the rotting usb ssd's affair )
mircea_popescu: so then as a factual matter, if asciilifeform threw out erry box that rebooted "by itself" for no apparent reason, pizarro would be missing uy1
asciilifeform: but indeed i'd much like to move to a 'near-errybody on rockchips' , ~these~ can approximate the ideal of 'treat irons as toilet paper, discard on 1st sign of rot'
asciilifeform: this is a case where '1024 chickens' wins
mircea_popescu: i confess i have nfi what makes you think commodity hardware failed in this case.
asciilifeform: i dun currently have any notion of knowing what, precisely , failed
mircea_popescu: outside of hard drives, and capacitors on OVER FIFTEEN YEAR OLD motherbopards, i have not witnessed this wonder myself, of failing hardware.
mircea_popescu: kudos to you, but nevertheless.
asciilifeform: ram, typically
asciilifeform: ( why -- i do not know. but ram appears to age, possibly ion migration )
mircea_popescu: but the ram in that box is as new as a kitten.
asciilifeform: the only new iron in the cage is the rk's.
BingoBoingo: The solid state drives in UY1 were new
mircea_popescu: i'm confused, you bought used ram ? i seem to recall a discussion...
asciilifeform: and disks yes
asciilifeform: mircea_popescu: 100% of the x86 iron in the cage is 2009-11 vintage.
BingoBoingo: From what I understand the ram came with the chassis
mircea_popescu: weren't you shipping a bunch of new rams to make it ?!
asciilifeform: ( can't speak for colo subscribers such as trinque , referring to pizarro irons )
asciilifeform: mircea_popescu: disks
mircea_popescu: i guess i must've mixed things.
asciilifeform: ssds are new, no one buys used ssd, it'd be like buying used toiletpaper
mircea_popescu: but i mean... so you have ten year old gb ram ?
asciilifeform: and 10yo caps etc
mircea_popescu: i never saw a mobo bust a cap and then boot by itself again tho. besides, he'd see a busted cap i imagine.
BingoBoingo: I saw no burst caps
asciilifeform: << typical example of when caps.
asciilifeform: they aint always visibly burst, but i'm inclined to think this wasn't the caps.
asciilifeform: ( or it wouldn't come back up. )
mircea_popescu: so to get this straight, your "most likely explanation" points to... ram failure resulting in kernel panic... twice ?
mircea_popescu: (i think the third reboot was actually you guise, or not ?)
asciilifeform: mircea_popescu: jury's still out
asciilifeform: mircea_popescu: 3 unsanctioned, 1 planned reset
mircea_popescu: always gotta have a mle, jury can't be out
asciilifeform: imho mle is still thermals.
mircea_popescu: ie, a fan stoppedf by itself, and then started working again, by itself ?
mircea_popescu: THRICE ?
asciilifeform: not necessarily stopped, but obstructed
asciilifeform: mircea_popescu: keep in mind this was a ben_vulpes-baked box, i never saw inside of it ( dulap-III, dulap-spare, s-mg, s-mg-spare -- i cleaned with own hands )
mircea_popescu: indulge me. so the theory goes that an event with a probability inferior to 1e-4 / day occured three times in two days ?
← 2019-01-13 | 2019-01-15 →