log☇︎
80200+ entries in 0.538s
a111: Logged on 2016-11-19 19:31 Framedragger: but yeah, i've noticed that sshd works just fine (incl accepting new connections) even if cpu at ~100% and/or no free disk space. there's that.
Framedragger: anyway, i imagine a bunch of d(a)emons fighting for i/o... and yeah no one is saying that there's a straightforward solution...
mircea_popescu: i dunno that alf is a db engineer by trade, so it's entirely possible specific measures could help, especially if they're of the magic number ilk of "set X to Y in config file, we didn't docum,ent this anywhere but its tru!"
mircea_popescu: http://btcbase.org/log/2016-11-19#1570931 << and if there's two* i meant lol. ☝︎
a111: Logged on 2016-11-17 14:43 Framedragger: (and on a side note, "hang out on #trilema after splitting with gf" has been one of the more constructive choices i've made in my life)
Framedragger: i do need to try to V...
Framedragger: yah but there's also enough of stuff to be done :) if trinque does WoT browser i won't do in parallel just so that there are two
mircea_popescu: http://btcbase.org/log/2016-11-19#1570790 << amusingly enough, this is actually true. "i would like to lecture these monkeys in modern psychology" "ok ?" "can you design a shitproof semipermeable membrane that still allows my precious words to reach them ?" "uh. not really. whatever the fuck it is, sooner or later the shit will clog it" ☝︎
Framedragger: sure. "javascript in the backend", etc etc etc.; i mean, it's still a horrible language. but yeah.
mircea_popescu: tbh i think javascript is not ever hated in the context of its original domain, "make drawn man move his arms" ; it's whenever it tries to be other things that it draws ire.
a111: Logged on 2016-11-19 17:24 mod6: Framedragger: so with regard of the wotperson to wotperson (http://www.btcalpha.com/wot/trust/?from=mod6&to=mircea_popescu) and all ratings to wotperson (http://www.btcalpha.com/wot/user/mod6/); I think we liked those very much as they were. Of course any sort of improvements could be added if they make sense, etc.
Framedragger: re. JS, yeah makes sense. and http://btcbase.org/log/2016-11-19#1570796 << thanks for the feedback! (and i agree regarding comments, preferences make sense) ☝︎
a111: Logged on 2016-11-19 16:57 Framedragger: re. visualization, i like stuff like this (mouse over on labels around the circle), but it's a hella lot of JS, and i share the hate towards the latter: http://bl.ocks.org/mbostock/7607999 - what's nice about btcalpha visualization is that it uses by-now standard html5 canvas directives (<path>) with no need for JS.
a111: Logged on 2016-11-19 16:45 Framedragger: i suppose the idea could be to re-implement that, but using deedbot's view of WoT, and add additional things as desired.
asciilifeform: (and now i must surely bbl, this thread could go on and on.)
mircea_popescu: i'm not for a second saying HE could have done much better / differently.
asciilifeform: and mircea_popescu has point. i'ma back to wurk, bbl.
Framedragger: right, fair enough i guess.
asciilifeform: Framedragger: i would keep it personally if i had where. (see old thread.)
asciilifeform: and i will add that , aside from the login keyz, there are 0 secrets on the box.
Framedragger: asciilifeform: then, i think, box needs to physically reside with someone in your WoT. i'm just saying. maybe cheap slippery slope sophism.
mircea_popescu: asciilifeform you gotta learn to judiciously allocate your time ; Framedragger the problem isn't so much locking up the box - it's that the box will fuck up, this being "foss" bullshit, and then the owner will be like "i dun wanna pay for a box i can't use, notwithstanding this is what i claimed i want".
Framedragger: (this is a case where i am more cynical than asciilifeform: a DC you don't know, what illusions does one have; okay, one could install a sensor which signals if box was 'tampered with'; they could still take box offline, clone memory, set up a replica. etc etc etc.)
asciilifeform: physically it ~can~ dump ram, research proj i led at the time investigated how to coax it to do this, among other things
asciilifeform: i had occasion to install such a machine a few yrs ago, somewhere, it came with 'kvm chipset', but this feature was not advertised. the advertised feature is just that, kvm, where you get a tty prompt.)
asciilifeform: well the one i subscribed to in the past, had a physical cart, with wheels
asciilifeform: i only know how mine work.
mircea_popescu: uh. i still wonder how you think computers work.
asciilifeform: i dun recall specifying 'remote admin card'
asciilifeform: i had nfi that we were even buying a 'fritz chipped' box.
a111: Logged on 2016-11-19 16:16 asciilifeform: i wrote, in the qntra piece, 'examine debianized boxes for nsaware'. now 'owner' will have a chance to clean up before any mass 'examination' takes place.
asciilifeform: incidentally i have machines right here in my house subjected to far harsher loads , 24/7/365, and somehow -- mysteriously -- they never ONCE suffer 'martian' failures.
Framedragger: i'm not sure of course.
Framedragger: by the way, i haven't ever used it, but from reading around it appears that streaming replication may indeed be quite efficient. every time row is inserted, row is sent off to remote replica. but this does not really require cpu. so maybe it wouldn't slow things down further / wouldn't be particularly slow even if db being clobbered 24/7
asciilifeform: ever since the theft of the original phuctor machine, i keep scrolling logs on SEPARATE LCD in real time, 24/7/365
Framedragger: (i remember having use from a super stupid `free > memory.log` cron job every minute or so)
Framedragger: but yeah, i've noticed that sshd works just fine (incl accepting new connections) even if cpu at ~100% and/or no free disk space. there's that. ☟︎
mircea_popescu: i dunno that they actualy bother keeping it for client managed boxes.
asciilifeform: i had a RUNNING SESSION 24/7 on own display, and it ground to a half
asciilifeform: i'm still convinced re syn flood.
mircea_popescu: though i must confess syn flood had a better ring to it altogether.
asciilifeform: let'em reboot. and i will start on the sawing-apart of the werker and wwwtron as soon as my hands are again free.
asciilifeform: the code has guaranteed run bounds. so i cannot make any comment re 'in general', there is nothing to fix.
mircea_popescu: asciilifeform in point of fact i have asked the public force to locate car drunk ho abandoned "dunno where". they did. i didn't throw a shitfit about "o noes surveillance state". this because a) i asked them to do it and b) obviously they just put the number in the stolen cars interface and then a patrol saw it.
mircea_popescu: sometimes i wonder how you think computers work.
mircea_popescu: well, until you insisted i ask, nobody did. once i ask, they gotta do specific things.
asciilifeform: but i dunno that anyone other than mircea_popescu had any business knowing that the process were called.
asciilifeform: also i will add, 'werker' saturates 15/16 cores for 20 MINUTES of any given runtime.
mircea_popescu: anyway, make a call, do i have it rebooted or let it be for a while and see if it digs itself back out ?
asciilifeform: ( i certainly do not recall giving the monkey an account on the box )
asciilifeform: i'd like to know!
Framedragger: (how can they even see the name? kvm? i thought it was baremetal?)
asciilifeform: i have nfi. but process runs reniced.
mircea_popescu: and unless maz actuallty reads the logs instead of doing his work, I TOLD YOU SO
asciilifeform: supposing i even see it again.
Framedragger: asciilifeform: of course. and also don't treat this as high priority, i prolly wont look at it before wednesday anyway. but would be interesting to take a look - and i'd take a look
asciilifeform: Framedragger: you are welcome to a copy of db as soon as i get my hands on it again
Framedragger: i'm thinking whether it'd be worth it to just have a static replica of db-as-it-currently-is, for now. as in for "i want to touch data, there's an outdated html file on loper-os i guess?" cases.
Framedragger: i'm sorry but you didn't convince me in regards to the 'amount' of data. > 100mil row postgres with > 100 gb of data in a 8GB ram server ran fine. and while phuctor may be a more demanding beast, shouting '5 mil keys, MILLION!' doesn't convince
Framedragger: i have no doubt that a sillicon valley version would be a datacentre of mongodb nodes which constantly fail and corrupt data, and cost millions. :)
asciilifeform: Framedragger: you do not seem to understand, so i will give illustration: http://www.swapmeetdave.com/Humor/Workshop/Lumber-Car-A.jpg
Framedragger: quite sure the diff'ing / updates were thought out thoroughly, i.e. time complexity is constant.
mircea_popescu: "oh it's on phuctor and why do i need to do anything", which is how we got "those debian experts are flyeyeing the code so why should i have a clue".
mircea_popescu: "oh maybe i have a page from 2005 somewhere"
Framedragger: by 'it', i mean scriba submitting to archive.is
Framedragger: mircea_popescu: right, will get it done. i had started on it, got sidetracked by the python encoding problem, and the got sidetracked by other stuff. need to re-trace, and will first do the archival bit.
Framedragger: (i thought someone was supposed to retroactively archive all links in all logs of all times?)
mircea_popescu: Framedragger your scriba could crawl all links in chan and archive them. phf said a while back he almost has this but i've yet to see it
asciilifeform: Framedragger: the fundamental problem is that i will have to rewrite WHOLE THING for any of this to possibly happen.
Framedragger: asciilifeform: (instead of doing a more obscure "query-able phuctor-and-stuff db" thing i could help with some kind of phuctor-public-display-vps infrastructure setup. just sayin'. thing's not clear in my head.)
asciilifeform: i dunno that anyone other than google and yandex ever made a ~full phuctor~ snapshot.
mircea_popescu: i.... think not.
mircea_popescu: in other wtf : all https://archive.is/http://phuctor.nosuchlabs.com/phuctored are from the same one day in may ; while all https://archive.is/http://phuctor.nosuchlabs.com/ are from the same one day i nseptember.
Framedragger: i've read, i am doubly interested due to vagueness of said broadcasts :p
Framedragger: a couple of vps providers i used have nice APIs
Framedragger: with regards to vps i could help, if help/hands are needed. i know you have other priority stuff asciilifeform. also, don't know what the meta-priority level here is. (i.e., compared to other projects etc)
asciilifeform: i really ought to have made entire thing a c proggy that shits out static html once in a while.
a111: Logged on 2016-11-19 15:52 asciilifeform: the folx whose chumpnet we blew open, with the debian boxes, i suspect -- trying to make their displeasure known.
Framedragger: even if more than that - all of that shit can be cached, static html pages. maybe i'm oversimplifying.
Framedragger: i don't know why you need true real time, tbh.
Framedragger: i'm not saying that going for cheapest ad-hoc option is accetabru. just, a display box showing static content needs much less.
Framedragger: asciilifeform: wouldn't be sure re. costs. vps can be $5/$10 a month, and stuff i used for ssh key crawling (scaleway) can bill hourly
mircea_popescu: nono i was being metaphorical. exception in my head.
mircea_popescu: Framedragger i've not yet got through the log because parsing threw exception on that line i quoted :D
a111: Logged on 2016-11-19 16:52 Framedragger: asciilifeform: at least with public-static and phuctor boxes being separate, you'd have access to the latter if it were private. (but i guess you could object with "private/undisclosed box on the internet, what is this oxymoron!")
Framedragger: mircea_popescu: display/processing separation is why i mentioned http://btcbase.org/log/2016-11-19#1570743 ☝︎
asciilifeform: i dun think anything short of unplugging the box will stop processing.
Framedragger: i concur ^
asciilifeform: i will point out that, unless the box has been stolen, it is still processing keys at same rate
Framedragger: (by other people's sites i mean sites that i'm responsible for.)
Framedragger: because 1) other sites' experience may be impacted, and 2) phuctor db would place some load on things. why = because i'd create a few indices, those would hog some memory, and assuming users want to do quite a bit of sorting etc, would take some cpu time as well. just sayin'. nothing scientific.
a111: Logged on 2016-11-19 17:18 ben_vulpes: Framedragger: i am also curious to know what kind of requirements you have for a vps that your current loggotron doesn't serve.
Framedragger: http://btcbase.org/log/2016-11-19#1570793 << current loggotron also runs on vps, and in itself it requires very few resources. no db use, even. at this point there's a bunch of stuff and other people's sites running on that vps, i don't feel comfortable adding additional load. ☝︎
asciilifeform: the box per se is ultra-responsive, even when 'werker' is firing (i leave a core open)
mircea_popescu: i can tell you what wtf right now : learn not to overload box.
mircea_popescu: http://btcbase.org/log/2016-11-19#1570685 << it pings with ~2-300ms, however i see no web interface. ☝︎
trinque: that was me saying I'll do it.
mod6: <+trinque> Framedragger: ben_vulpes: it probably makes the most sense for me to do the WoT browser << perhaps. i'd be up for anything really at this point.
mod6: I think the idea would be to get back to some sort of analog of what mike_c had in place. And then add improvments as necessary.
mod6: Framedragger: I do like this graph for sure, http://bl.ocks.org/mbostock/7607999