log☇︎
507700+ entries in 0.341s
nubbins`: so this is all agreed, anything that can be modified should be signed
ascii_field: user is free to examine the (signed) update on his own time
nubbins`: there's gotta be a sane way to do this over the wire
mthreat: after the land of chipotle
nubbins`: the list'll be stale by the time someone gets a box
ascii_field: SIGNED update would have to be on attached storage
ascii_field: can't rely on anything on the net staying put
nubbins`: where's the seed list now, in a fucking .c file or something?
ascii_field: i suggested this, yes
nubbins`: gpg and curl, at the very least
ascii_field: and mircea_popescu will say 'bah, unreliable, you're driving up engineering cost 1000 times'
ascii_field: nubbins`: this means gpg on the nodes
nubbins`: in fact, i can almost certainly have this process be automatic
ascii_field: there is no other side to this
nubbins`: and is thus useless for inclusion in a baked-in
ascii_field: nubbins`: there you go
nubbins`: my ip address will be fine to use as a seed node, but it /may occasionally change/
ascii_field: or emplacing own turd there
ascii_field: one of the few things usg is reliably good at is making a particular ip unusable
ascii_field: turning an ip address into a critical constant in a field-deployed machine paints a big fat target on that ip for usg
ascii_field: so, what, we're expecting operators to throw out perfectly good boxes to add a fucking seed node!?!
ascii_field: nubbins`: this also.
ascii_field: likewise physical delivery is expensive. (witness kakobrekla's attempt to buy so much as one pogo and still have it 'feel disposable')
nubbins`: i think a lot of it is genuinely overstock too
ascii_field: (the vendor is selling a loss-leader, hoping to attract chumps to its 5/mo. 'cloud' crapolade)
ascii_field: right now some bezzlatronic idiot is paying the difference
ascii_field: once the twentybux boxes run out (and i expect that it will happen sooner rather than later) prepare to pay six times more for each node.
ascii_field: i utterly understand the desire to 'tivoize' these things so they don't get turned into phoundation nodes by lamers. but this can be done without idiot waste.
ascii_field: ben_vulpes: no, nodes don't get updated << so, let's go into this in detail. they ship with all the seed ips they will ever have? for the record i proclaim this to be retarded. -keys- are the constants of our world, not ip addrs
ascii_field: danielpbarron: i'd like more specific info on how to make my own u-boot << i have a u-boot built from official source. the buildroot thing -also- builds a u-boot, as a matter of course, but i have not yet tested that one
assbot: Bubble talk - Sam Altman ... ( http://bit.ly/1BnuE6y )
punkman: http://blog.samaltman.com/bubble-talk "This bet is open to the first VC who would like to take it . The loser donates $100,000 to a charity of the winner’s choice." meh
mircea_popescu: nubbins` installs the dix first!
nubbins`: (who in the fuck installs the docs /first/?)
nubbins`: might be easier than figuring out why the docs won't install
nubbins`: you could remove install_docs from that line if you wanted
nubbins`: here's the Makefile from openssl
nubbins`: if you're only doing the first step, ourlibs won't be fully populated
nubbins`: ourlibs is where things get put as they're compiled
danielpbarron: so i think it isn't finishing
danielpbarron: in my "ourlibs" directory, it looks like there are only doc files
nubbins`: if you try running a 64bit program on a 32bit os, you're gonna be sad
mircea_popescu: actually... maybe try 16 bit something ?
nubbins`: match the os
danielpbarron: oh also, should i be using the patched for 32 auto.sh if my cpu is 64 but my OS is 32 ?
nubbins`: while docs generally aren't crucial, i wonder if that stops some later process from happening
mircea_popescu: https://www.youtube.com/watch?v=ge0WaPvECPw << there, sung.
danielpbarron: on the openssl part, it ends with this: Makefile:640: recipe for target 'install_docs' failed
nubbins`: there's no beef in the north
nubbins`: ^ rabbit hole if anyone's into reading dead trees
nubbins`: hey, the best swordsmaster who ever lived was bested by a farmer with a quarterstaff
ben_vulpes: swinging sticks at each other in the muddy courtyard
nubbins`: just wetting the stones
thestringpuller: so we are building axes and swords today? :D
nubbins`: yeah, find is one thing, build properly is another
danielpbarron: it seems to find all three of openss, boosts, and db
nubbins`: it'll be easier to spot which builds fail
nubbins`: then comment out that one, uncomment the next, etc
thestringpuller: danielpbarron: i've done the same thing tho. i only had access to 32bit windows for my gaming rig, and i have to wipe the entire thing to install a 64bit version.
nubbins`: it'll be a bit of a pain, but if you can comment out your auto.sh so that it only builds one component
nubbins`: <+danielpbarron> this is a different one than the one that didn't have the db_cxx.h <<< ah
danielpbarron: (which is what i have been doing up until this point)
danielpbarron: this is a different one than the one that didn't have the db_cxx.h
ben_vulpes: you'll see that if you run ./auto.sh more than once, non?
mircea_popescu: danielpbarron don't worry, you also put 64 bit on the 32 bit arm.
nubbins`: try that and paste the result
danielpbarron: so i'm the dummy that didn't realize he had a 64 bit cpu and put 32 bit OS on it
nubbins`: haha actually sorry, the pastebin i linked above has 1 error
danielpbarron: i think my laptop has a 64 bit cpu
nubbins`: save that as autobdb.sh
nubbins`: try this: http://pastebin.com/raw.php?i=Ph1KJm0W
nubbins`: still failed 1 target above
danielpbarron: this is output when i use the 0.5.3.1 release and the patched for 32 auto.sh
danielpbarron: oh weird, it does exist in there when i try with the prepatched 0.5.3.1
nubbins`: then run it, see what errors it throws
nubbins`: comment out everything in auto.sh but the chunk where it builds bdb
danielpbarron: i guess that means db isn't finishing
nubbins`: working backwards, next step is to modify auto.sh
nubbins`: okay, so that's why that step can't find it :D
nubbins`: danielpbarron at the risk of being silly, does db_cxx.h exist in /home/dpb/pogo/bitcoind/static.7th/ourlibs/include
danielpbarron: ...failed updating 1 target... << this happens when i use the prepatched 0.5.3.1, the looky loo outputs correct paths
danielpbarron: that's the one i'm probably using already
danielpbarron: yeah no, i mean how to make my own u-boot
assbot: Qui's techNOLOGY Blog: Hacking the Pogoplug v4 (Series 4 and Mobile) with Linux (Debian or Arch) ... ( http://bit.ly/1zZyYqB )
danielpbarron: my adulterated pogo behaves strangly, sometimes it uses stock u-boot and sometimes it uses some thing i flashed to it, and i'm not sure why it changes
danielpbarron: i'm not sure what asciilifeform recommends for this
danielpbarron: on a related note, i'd like more specific info on how to make my own u-boot
assbot: Logged on 25-03-2015 04:13:16; asciilifeform: (it croaks, having no meaningful rootfs, but that's for tomorrow)
danielpbarron: http://log.bitcoin-assets.com/?date=25-03-2015#1069675 << ---[ end Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) ☝︎
mircea_popescu: moldavian count. famous pop singer of the 20s. after which, took up clay pigeon shooting, and became world champion 4ever.
assbot: Manole Stroici - Și pe tine or să te doară... - YouTube ... ( http://bit.ly/1CP3nOP )
mircea_popescu: https://www.youtube.com/watch?v=umeS2PWcRU0 << this guy.
nubbins`: instead of just mousing around the login screen with no kbd input :D
mircea_popescu: nubbins` cool, that way you can play too :D
nubbins`: mircea_popescu hey who knows, maybe there'll be an osx build some time this summer
jurov: okay, but be sure to implement faster deposits :>
mircea_popescu: i'm so fucking excited about this eulora release...
asciilifeform caught up with logs, off to meatspace for a spell
danielpbarron: yeah, i think i remember doing that the last time i went through all this
jurov: fatal error: db_cxx.h: No such file << that's known issue on gentoo, you have to add -I <bdb path>