jbenet changed the topic of #ipfs to: IPFS - InterPlanetary File System - https://github.com/jbenet/ipfs -- channel logged at https://botbot.me/freenode/ipfs/ -- code of conduct at https://github.com/ipfs/community/blob/master/code-of-conduct.md -- sprint at https://github.com/ipfs/pm/issues/7
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
neoteo has joined #ipfs
neoteo has quit [Ping timeout: 244 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
anshukla has joined #ipfs
Evermore has joined #ipfs
<Evermore> I know you guys have been discussing something like this, but it didn't really hit me until now: With the FUSE mount, can't I launch an application right from IPFS?
<Evermore> Makes me wish it worked on Windows, for games
<whyrusleeping> Evermore: yeah, you can launch an application right out of windows
<whyrusleeping> but
<whyrusleeping> er
<whyrusleeping> s/windows/fuse/
<whyrusleeping> but, the +x bit wont be set
<whyrusleeping> we are working on adding unix permissions to our files, as well as metadata
<Evermore> that seems like a sticking point
<Evermore> ok
<whyrusleeping> yeah...
<whyrusleeping> you could do "python /ipfs/<hash>"
<Evermore> Well a lot of programs are C++ or have C++ components
<Evermore> or libraries
<whyrusleeping> zero install python packages and libraries
<Evermore> my* programs
<whyrusleeping> within a month we should have this fixed
<Evermore> cool
<whyrusleeping> yeap! i think so
orzo has quit [Ping timeout: 276 seconds]
<Evermore> If only I could convince other people to do that, so I don't have to find a directory to save their games in
lgierth has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has quit [Quit: Leaving.]
wallet42 has joined #ipfs
<ipfsbot> [go-ipfs] whyrusleeping pushed 1 new commit to feat/bitswap-speed: http://git.io/vkeWY
<ipfsbot> go-ipfs/feat/bitswap-speed fdb94c6 Jeromy: adjust naming
<whyrusleeping> Evermore: you could just have saves be in their ipns directory
orzo has joined #ipfs
<Evermore> whyrusleeping: Could do that. They'd still have to pick a unique name
<whyrusleeping> Evermore: would they?
<Evermore> I guess I could use a launcher that changes PWD and HOME to the ipns directory for that program's hash
<Evermore> the Pandora has something like that for its PND application bundles
<Evermore> But I don't see how it would work normally
orzo has quit [Ping timeout: 272 seconds]
<whyrusleeping> Evermore: it could just be /local/ipns/.special-game-dir/
<Evermore> Yeah so .special-game-dir has to be unique to that author, right?
<whyrusleeping> yeah
<whyrusleeping> which, is normally fine to assume
<whyrusleeping> like .vimrc
<whyrusleeping> .bashrc
<whyrusleeping> things like that
<whyrusleeping> .your-application-name
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
orzo has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
guest7561 has joined #ipfs
guest756 has quit [Ping timeout: 272 seconds]
guest756 has joined #ipfs
guest7561 has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
robsyme has joined #ipfs
guest7561 has joined #ipfs
guest756 has quit [Ping timeout: 250 seconds]
nessence has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
anshukla has quit [Remote host closed the connection]
neoteo has joined #ipfs
guest7561 has quit [Ping timeout: 245 seconds]
neoteo has quit [Ping timeout: 245 seconds]
nessence has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
EricJ2190 has quit [Ping timeout: 264 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
lgierth has quit [Quit: Ex-Chat]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
pfraze has joined #ipfs
wallet42 has quit [Quit: Leaving.]
wallet42 has joined #ipfs
wallet42 has quit [Ping timeout: 258 seconds]
patcon|afk has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
neoteo has joined #ipfs
mquandalle has quit [Quit: Connection closed for inactivity]
neoteo has quit [Ping timeout: 264 seconds]
guest756 has joined #ipfs
Wallacoloo has quit [Quit: Leaving.]
kyledrake has joined #ipfs
<kyledrake> 'evening. I noticed that the config datastore is set to leveldb. Do I need to change that, or is this separate from the file storage?
<kyledrake> ah, the blocks are different, ic.
pfraze has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
inconshreveable has quit [Remote host closed the connection]
chriscool has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Wallacoloo has joined #ipfs
patcon|afk has quit [Ping timeout: 265 seconds]
patcon has joined #ipfs
sharky has quit [Ping timeout: 276 seconds]
guest756 has quit [Ping timeout: 240 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
sharky has joined #ipfs
chriscool has quit [Ping timeout: 246 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
inconshreveable has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
neoteo has joined #ipfs
neoteo has quit [Ping timeout: 250 seconds]
flugsio has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
robsyme has quit [Ping timeout: 258 seconds]
neoteo has joined #ipfs
mildred has joined #ipfs
wallet42 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
kyledrake has quit [Ping timeout: 258 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
guest756 has joined #ipfs
wallet42 has quit [Quit: Leaving.]
guest756 has quit [Ping timeout: 256 seconds]
timgws has joined #ipfs
timgws has quit [Changing host]
timgws has joined #ipfs
wallet42 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
timgws has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
rht_ has quit [Quit: Connection closed for inactivity]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
henq has quit [Quit: henq]
timgws has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
octalberry has joined #ipfs
henq has joined #ipfs
henq has quit [Ping timeout: 276 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
timgws has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<cryptix> gmorning
hellertime has joined #ipfs
atomotic has joined #ipfs
<wallet42> can I get the seeder count for a ipfs block?
atomotic has quit [Client Quit]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
lgierth has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<doublec> wallet42: I believe "ipfs dht findprovs <hash>" lists the nodes seeding that hash
<wallet42> thx
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Wallacoloo has quit [Ping timeout: 265 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
mildred has quit [Ping timeout: 276 seconds]
wallet42 has quit [Quit: Leaving.]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
mildred has joined #ipfs
saebekassebil has joined #ipfs
robsyme has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
inconshreveable has quit [Remote host closed the connection]
therealplato has quit [Ping timeout: 245 seconds]
nessence has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
nessence has quit [Remote host closed the connection]
nessence has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
daviddias has quit [Quit: Textual IRC Client: www.textualapp.com]
nessence has quit [Remote host closed the connection]
saebekassebil has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
saebekassebil has joined #ipfs
saebekassebil has quit [Client Quit]
daviddias has joined #ipfs
atomotic has joined #ipfs
<ipfsbot> [go-ipfs] diasdavid opened pull request #1292: Change one of the "ipfs" to "ipns" (master...patch-1) http://git.io/vkfDZ
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ekarlso has quit [Changing host]
ekarlso has joined #ipfs
domanic has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has joined #ipfs
<jbenet> Bye bye berlin <3 see you in august
<cryptix> jbenet: :) good flight
infinity0 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has quit [Quit: Leaving.]
wallet42 has joined #ipfs
therealplato has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has quit [Quit: Leaving.]
nessence has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
nessence has quit [Remote host closed the connection]
<ehd> jbenet: it was great having you :)
<ehd> *and all other p2pers who came to town. fantastic weekend :)
<ehd> cjb: yo, are you on twitter by the way? i'm @evilhackerdude
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<ipfsbot> [go-ipfs] rht opened pull request #1293: Add test for cat error message when `ipfs init` hasn't been run (master...cleanup-cat-error) http://git.io/vkJzE
<ipfsbot> [go-ipfs] rht opened pull request #1294: Cleanup tour (master...cleanup-tour) http://git.io/vkJzx
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
therealplato has quit [Ping timeout: 255 seconds]
mildred has quit [Ping timeout: 276 seconds]
therealplato has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<cryptix> ehd: cjbprime i think
atomotic has quit [Quit: Textual IRC Client: www.textualapp.com]
pfraze has joined #ipfs
rht_ has joined #ipfs
<whyrusleeping> anyone know of any real low power linux boxes?
<whyrusleeping> the rPi uses ~270mah
<whyrusleeping> which is somewhat hefty
<whyrusleeping> if i could get something <100ma that would be great
<headbite> I just watched a video on that tegra K1 dev board.... it's supposed to idle pretty low but it's like 200 bucks from what I remember, and it's kind of large.
<headbite> their running some modded ubuntu 14.04
<ipfsbot> [go-ipfs] whyrusleeping pushed 2 new commits to master: http://git.io/vkJhO
<ipfsbot> go-ipfs/master b2e45da David Dias: Change one of the "ipfs" to "ipns"
<ipfsbot> go-ipfs/master b3e73c6 Jeromy Johnson: Merge pull request #1292 from diasdavid/patch-1...
octalberry has quit [Ping timeout: 276 seconds]
nessence has joined #ipfs
octalberry has joined #ipfs
<whyrusleeping> headbite: yeah, the tegra k1 is huge, and has some pretty hefty power draw
nessence has quit [Read error: Connection reset by peer]
octalberry has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping> oooh, but it idles low
<whyrusleeping> youre right
<headbite> yeah.... and all those graphics cores might be useless depending on what you're doing.
<whyrusleeping> yeah, the graphics cores are definitely useless, lol
<whyrusleeping> i'm working on making a traveling ipfs node
<whyrusleeping> to run in my backpack on a portable battery
<whyrusleeping> requirements are wifi, bluetooth, and a couple USBs
guest756 has joined #ipfs
<headbite> odroid has a few different pi type boards... they have those small ass clip on (emmc) ssd drives too. I don't know which one is best for power though.
<whyrusleeping> ooooh, ill take a look there
<whyrusleeping> i know i looked at the c1 a while back
<whyrusleeping> but i wasnt interested in power draw then
<headbite> the u3 has that emmc thing for sure...
<whyrusleeping> oooh, odroid q
<whyrusleeping> w*
<headbite> o yeah... that W looks hella small. LOL
saebekassebil has joined #ipfs
<headbite> their email support is pretty good.... they got back within a day given the time zone difference.
wallet42 has joined #ipfs
<whyrusleeping> nice
<whyrusleeping> the w looks perfect...
<headbite> looks like it's got the emmc on the bottom too
<whyrusleeping> schweet...
<whyrusleeping> and a lipoly battery connector
saebekassebil has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
bnvk has joined #ipfs
<whyrusleeping> headbite: the intel edison
<whyrusleeping> people are reporting an idle power draw of 40ma
<whyrusleeping> and its a dual core atom
<bnvk> hey y'all. was cool meeting the ppl from the team in Berlin last night!
<whyrusleeping> bnvk: wish i coulda been there!
<headbite> whyrusleeping: yeah... that looks interesting too. Damn it, now I want one. LOL
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
guest756 has quit [Ping timeout: 276 seconds]
domanic has quit [Ping timeout: 256 seconds]
patcon has quit [Ping timeout: 252 seconds]
guest756 has joined #ipfs
guest756 has quit [Ping timeout: 272 seconds]
wallet42 has quit [Quit: Leaving.]
wallet42 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
robsyme has quit [Read error: Connection reset by peer]
wallet42 has quit [Quit: Leaving.]
pfraze has quit [Remote host closed the connection]
patcon has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping> headbite: my problem with it, is that even though the board itself is only $50
<whyrusleeping> you have to buy all sorts of other things to do anything with it
pfraze has joined #ipfs
<whyrusleeping> cryptix: 1150 should be fixed
<headbite> yeah... I picked up a pair of odroid u3's and they were right around 200 bucks each when it was all said and done.
<whyrusleeping> er, s/1150/1050/
<headbite> I was going to use them as my daily web surfing machines but then I got this chromebook and those things have just been collecting dust. I should at least set them up as some bad ass motion capture cameras or something.
<whyrusleeping> i love chromebooks
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping> spent two and a half years using one as my primary dev station
<whyrusleeping> but once it got to the point where the ipfs test suite took 20 mins to run on it, i had to get something a little bit nicer
wallet42 has joined #ipfs
<headbite> The ecs liva is back down to 100 bucks. I've been looking at it more for my friends when their old desktops die. www.newegg.com/Product/Product.aspx?Item=N82E16856501007
<rht_> but 2.5 years from 2.5 years ago, chromebook specs have since then been beefed up (> 1 tick-tock cycle)
<whyrusleeping> rht_: oh, i know, ive had four chromebooks
<whyrusleeping> i had the samsung chromebook 2 13" most recently
<whyrusleeping> +kernel patch to use all eight of the cores
<rht_> and still not enough?
<whyrusleeping> it just doesnt compare to an i7 at 3.0Ghz
<whyrusleeping> the biggest issue really, was the lack of a real SSD
<rht_> it looks riskier to me to do the kernel patch than to get an ssd
<rht_> (what do you mean by *real* ssd?)
inconshreveable has joined #ipfs
<whyrusleeping> i mean, not a 16GB microSD card, lol
<whyrusleeping> and yeah, the kernel patch was... fun
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Tv` has joined #ipfs
<rht_> the furthest I have been to the kernel-land was to underclock my laptop to save battery actually (or I don't think it even needs kernel patch, i just have to set some param from /sys/something)
<whyrusleeping> oh yeah, you just change the governor
<whyrusleeping> i remember a linux install i used once that had it as an option on the tray
<whyrusleeping> krl: i just noticed the weird output thing on dht findprovs
<rht_> (did the linux install has other options as well?)
<whyrusleeping> rht_: no, it was just a little drop down menu that let you select a cpu speed
tibor_ has joined #ipfs
<rht_> is it worth it to do so (cpufreq scaling) nowadays? or the answer depends on too many parameters. I never did a benchmark
<rht_> ah, anyways, I'm actually up for doing the decimate eventlog issue *now*
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping> okay, cool
tibor_ is now known as tibor
<whyrusleeping> so, it provides some moderately valuable information
<whyrusleeping> and jbenet wants to 'keep it around'
<whyrusleeping> but we definitely need to stop writing it to disk
<whyrusleeping> and it also needs to not format any arguments unless its being output somewhere
<whyrusleeping> our idea was that we can expose an eventlogs http endpoint, and while there is a client to that endpoint, eventlogs are written out through that
<rht_> (no benchmark here http://www.linux-phc.org/)
<rht_> so there won't be local logs? there is a todo comment that says to expose loglevel as an argument (wait a sec)
<whyrusleeping> yeah, i think eventlogs should be displayed at a certain log level too
<whyrusleeping> but they have so much extra info attached to them that its best to also have a separate way of acessing them
<rht_> 380 mb logs | tar with bz2 outputs a 15.5 mb file
<whyrusleeping> yeah... we also could be compressing, lol
<rht_> is this one reason why systemd uses binary logs?
<whyrusleeping> i assume so
<whyrusleeping> but systemd also has working log rotation...
<whyrusleeping> unlike ours
sharky has quit [Quit: Leaving]
sharky has joined #ipfs
<rht_> logrotate can be handy for long term nodes
<whyrusleeping> rht_: we *have* logrotation, supposedly
<whyrusleeping> but it doesnt work for some strange reason
<rht_> from logrus readme:
<rht_> #### Rotation
<rht_> Log rotation is not provided with Logrus. Log rotation should be done by an
<rht_> external program (like `logrotate(8)`) that can compress and delete old log
<rht_> entries. It should not be a feature of the application-level logger.
<rht_> so there are eventlog, go-logging, prefixlog, and logrus?
<whyrusleeping> rht_: :/ yeah... we have a lot of those
<whyrusleeping> but we do have a log rotator somewhere
<rht_> can't find it
<whyrusleeping> we use natefinch/lumberjack
<whyrusleeping> but youre right, it probably should be done outside of ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<rht_> yeah less moving parts to be maintained, more focus on optimizing git, bitswap perf
tilgovi has joined #ipfs
ei-slackbot-ipfs has quit [Remote host closed the connection]
ei-slackbot-ipfs has joined #ipfs
<whyrusleeping> yeah
<whyrusleeping> headbite: btw, the odroid w's are on clearance here: http://ameridroid.com/products/odroid-w-development-kit-as-is
<whyrusleeping> thats a pretty damn good deal
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has quit [Quit: Leaving.]
www1 has quit [Quit: Leaving.]
<ipfsbot> [go-ipfs] whyrusleeping pushed 1 new commit to feat/bitswap-speed: http://git.io/vkTuY
<ipfsbot> go-ipfs/feat/bitswap-speed 4ef8de5 Jeromy: clean up organization of receivemessage and fix race
<whyrusleeping> Tv`: i have a crazy idea to run by you
<Tv`> whee
<whyrusleeping> so, with flatfs, instead of writing each block as they come in each in its own file
<whyrusleeping> we write it all to some journal file, in serial
<whyrusleeping> then the write call can return
<whyrusleeping> and in the background we can break up the journal file into the actual files each block needs to go to
<whyrusleeping> (this would entail serializing all writes)
<whyrusleeping> i've been seeing weird performance with lots of parallel block writes
<Tv`> that's what i've been calling "arena storage"
<Tv`> it's actually better than breaking up in to smaller files
<Tv`> less overhead
<Tv`> but it's not really all that trivial, and you can get a lot of the benefits even without it
<whyrusleeping> true
<Tv`> e.g. if you have a concept of "a group of writes", you only need to do the dir fsyncs once at the end
<Tv`> all of this complicates apis ;)
<whyrusleeping> yeah, i
<whyrusleeping> i would love to have a transaction based datastore interface
<whyrusleeping> bolt does that really nicely
<crest> whyrusleeping: boltdb is optimal to shock unsuspecting sysadmins with >30GB VSZ
<whyrusleeping> crest: its all virtual :P
<crest> whyrusleeping: i know
<whyrusleeping> those sysadmins most likely have never run zfs either ;)
<crest> and it's file backed virtual memory
<crest> even if you use the memory mapped file it can be demand paged
<crest> but i was surprised by boltdbs address space usage the first time i installed consul
<whyrusleeping> yeah, its kinda funny that linux reports it as 'memory used'
<Tv`> virtual memory
<Tv`> back in 32-bit days, you could run out of that, it made a lot of sense to track it
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
mildred has joined #ipfs
wallet42 has joined #ipfs
Wallacoloo has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Wallacoloo has quit [Ping timeout: 272 seconds]
Blame has quit [Quit: Connection closed for inactivity]
sharky has quit [Quit: Leaving]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has quit [Quit: Leaving.]
mildred has quit [Quit: Leaving.]
mildred has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has joined #ipfs
wallet42 has quit [Client Quit]
sharky has joined #ipfs
daviddias has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
pfraze has quit [Remote host closed the connection]
kyledrake has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
atomotic has joined #ipfs
daviddias has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
patcon has quit [Ping timeout: 272 seconds]
anshukla has joined #ipfs
sharky has quit [Quit: Leaving]
inconshreveable has quit [Ping timeout: 265 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
sharky has joined #ipfs
mildred has quit [Ping timeout: 244 seconds]
mildred has joined #ipfs
<whyrusleeping> Tv`: where was that writeup for the pinning datastructure?
<whyrusleeping> nell: what was your website again?
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
inconshreveable has joined #ipfs
peerchemist has joined #ipfs
<whyrusleeping> Tv`: and thats not using the cdb stuff, is it?
<peerchemist> hello, I've just bumped into your project. Looks quite interesting so far. I have one question, and I cant find the answer on your website. Is it possible to have untrusted nodes? ie, is data readable to anyone on the grid?
<Tv`> whyrusleeping: last version is not
<whyrusleeping> peerchemist: data is readable to anyone on the network
<whyrusleeping> and its content addressed, so you can guarantee that the data you request is the data you get
<whyrusleeping> Tv`: alright, cool. was finding examples of databases built on ipfs/using ipfs as a database
atomotic has quit [Quit: Textual IRC Client: www.textualapp.com]
<peerchemist> @whyrusleeping, thx.
<whyrusleeping> peerchemist: but in the future, you will be able to keep data from being read by unwanted peers
<whyrusleeping> ipfs has many different layers, and they are designed to be swappable
<whyrusleeping> the system that does lookups for content is called the routing layer
<peerchemist> that is what I am interested in, having shared database but choosing who to trust
<peerchemist> that is how I got to your project :)
<whyrusleeping> so, there are two ways to do that
<whyrusleeping> one, you implement a different routing system that only lets your trusted peers access your content
<whyrusleeping> the other way, is to encrypt your content with a key shared between your trusted peers
<peerchemist> and those two work now? not feature request, but implemented?
<whyrusleeping> #1 is a feature request
<whyrusleeping> and #2 is coming fairly soon
<whyrusleeping> but you dont need to wait for us to implement it
<whyrusleeping> you can encrypt your data before adding it to ipfs
<whyrusleeping> i should say, a built in way for doing encryption is coming soon
<peerchemist> well that is what I am after actually, an automated way
<peerchemist> is the design modular, can I rip out web UI?
<whyrusleeping> everything is very modular
<whyrusleeping> you can rip out the webui and the CLI if you like
<whyrusleeping> you can even rip out the DHT (and replace it with something to do the same services)
<peerchemist> awesome, sounds better than Tahoe-Lafs already
<whyrusleeping> tahoe-lafs has slightly different use cases than us
<whyrusleeping> tahoe assumes you control all the nodes, and redundandtly stores the data on the other machines
<whyrusleeping> ipfs only puts data on peers that request it
<peerchemist> and than those nodes become storage nodes?
<whyrusleeping> until they remove the data from their node, yeah
<whyrusleeping> think of it like how torrents work
<peerchemist> yes
<peerchemist> I understand
<peerchemist> can I tell you about what I am after? It seems that you could be interested in the subject
<peerchemist> I am looking for a way to implement DAC, decentralized autonomous company
<peerchemist> where data would be shared among all nodes, but data is visible to only the nodes who are assigned with some "role"
<peerchemist> others are just support
<peerchemist> role is granted by the all participants, by majority vote
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
www has joined #ipfs
<whyrusleeping> hrm... thats an interesting problem
<whyrusleeping> the sharing problem has to be solved by encryption
<peerchemist> yes, that is how I got to Tahoe (as tahoe can do that)
<whyrusleeping> thats the only way you can get nodes to hold data, but not be able to read it
<peerchemist> but tahoe lacks some other stuff
<whyrusleeping> so, we are going to implement what we are calling a 'keystore' soon
<whyrusleeping> that will allow you to manage multiple keys
<whyrusleeping> and share them through the network with chosen peers
<peerchemist> will that be equivalent of Tahoe's permissions
<whyrusleeping> it can be used that way
<whyrusleeping> i think tahoe calls them 'caps' right?
<peerchemist> yes, capabilities
<whyrusleeping> once our keystore is implemented i'll do a writeup on how to do caps style stuff in ipfs
<peerchemist> okay, awesome
<peerchemist> and why am I in process of giving up on tahoe (and got here) is because tahoe is not modular
<peerchemist> and it will be difficult to integrate it with other software, also it is bloated
<peerchemist> so your project really does look interesting
mildred has quit [Quit: Leaving.]
<peerchemist> but I need some more reading about ipfs
<whyrusleeping> well, i suggest the whitepaper, and also http://github.com/ipfs/faq
<whyrusleeping> and there should always be someone around in this channel to answer your questions
<whyrusleeping> i have this odd habit of being unconcious and not on IRC for about a third of my day every night
<peerchemist> okay, thanks. Will start reading and see if this fits into my idea
<peerchemist> what is use case you guys are aiming at?
EricJ2190 has joined #ipfs
<whyrusleeping> there are a lot of use cases we want to happen
<whyrusleeping> thats a hard question to give a single answer to
<whyrusleeping> i want to have a package manager built on ipfs
<whyrusleeping> websites, databases
<whyrusleeping> the distributed web!
<peerchemist> yes, I understand that potential is huge
<peerchemist> well for me it is this with decentralized database powering a true autonomous company
Wallacoloo has joined #ipfs
<peerchemist> do you have any benchmarks?
<whyrusleeping> peerchemist: we have some benchmarks of various internal components
<whyrusleeping> what numbers are you interested in?
Wallacoloo has quit [Ping timeout: 250 seconds]
<peerchemist> mostly latency
notduncansmith has joined #ipfs
<peerchemist> not so much into throughput
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping> so, latency on a warmed up node is going to be 1RTT in the best case, O(log(network size)) RTT's in the worst case, and on average, if the data exists in the network, it shouldnt take more than 3 RTTs to complete your request
<whyrusleeping> the best case being, we have an open conection already to the peer with the data we want
<peerchemist> RTT is round trip?
<whyrusleeping> yeah, round trip time
<peerchemist> IPFS still needs original source for peer discovery?
<peerchemist> like torrent
<whyrusleeping> peerchemist: kinda
<whyrusleeping> so, we prefill the default config file with nodes that we run
<whyrusleeping> so people can use those for bootstrapping
<whyrusleeping> but! if you have other peers on your network, we have local peer discovery via mdns
<whyrusleeping> Tv`: 192.168.0.0/12 would look like 11111111 11110000 000...
<whyrusleeping> what byte order is that?
ryepdx has quit [Ping timeout: 244 seconds]
<whyrusleeping> is the second number 15? or 240?
<whyrusleeping> (or anyone else who understands ip, just assumed that tv would know)
<lgierth> pretty sure that 11111111 is 255
<whyrusleeping> yeah, i'm confident on that one
<lgierth> ah. got ya
Blame has joined #ipfs
pfraze_ has joined #ipfs
<whyrusleeping> its 11110000 that i'm not sure about
<whyrusleeping> for ip masking
<grawity> it's 255.240.0.0 definitely
<whyrusleeping> grawity: awesome, thanks!
ryepdx has joined #ipfs
<whyrusleeping> thats what i thought at first, until i started writing the bitmasking loop and confused myself greatly
<grawity> though RFC 1918 only reserves 192.168.0.0/16
<grawity> so /12 would kinda step on other organizations' addresses
<gwillen> yes, what grawity said -- the correct bitmask for 192.168 is /16 or longer
<grawity> I see CIDRMask() in https://golang.org/src/net/ip.go if that's useful
<whyrusleeping> gwillen: yeah
notduncansmith has joined #ipfs
<peerchemist> one more question. Is there an RPC to communicate with daemon?
notduncansmith has quit [Read error: Connection reset by peer]
<peerchemist> what if I want to run daemon as another sandboxed user (for security)
kyledrake has quit [Ping timeout: 256 seconds]
anshukla has quit [Ping timeout: 250 seconds]
<whyrusleeping> peerchemist: currently daemon communication happens via http on the localhost
<whyrusleeping> if you want to control it remotely you can put some sort of auth in front of it (using nginx magic)
inconshreveable has quit [Remote host closed the connection]
inconshreveable has joined #ipfs
<peerchemist> I think you could benefit greatly by having an RPC protocol implemented. Doing so you will allow users to run IPFS daemon in sandbox (away from touchy data) and control it via command line with ease
<whyrusleeping> oh, we have RPC
<peerchemist> :)
<whyrusleeping> its http
<whyrusleeping> and the CLI uses it to communicate with the daemon locally
<peerchemist> that is what I am asking you :)
<whyrusleeping> but we by default (for security reasons) restrict the rpc listener to listen only on the localhost
<whyrusleeping> if you want to use it remotely (over the open web) it would be best to put some sort of auth layer in front of the RPC endpoint
anshukla has joined #ipfs
<peerchemist> localhost is enough
<whyrusleeping> then yeah, we have what you ask :)
<peerchemist> is just running a node helping a network in any way?
<whyrusleeping> yeap! you strengthen the dht
<peerchemist> then you can expect a node soon
<whyrusleeping> :D
<peerchemist> 0-24/7
<peerchemist> but I need to secure the daemon first
<peerchemist> some systemd magic
<whyrusleeping> oh, yay systemd, lol
<grawity> that's mostly just mount namespaces and some prctl()
therealplato has quit [Ping timeout: 265 seconds]
<peerchemist> systemd rocks
<whyrusleeping> there was an init.d script laying around somewhere...
<Tv`> whyrusleeping: /12 means 12 bits set at front
<Tv`> in the same order as the octets you listed, 192.168.0.0...
<peerchemist> I will push systemd to my git hub tomorrow probalby
<peerchemist> btw, I got a lot of this now while testing: >> 00:01:21.322 ERROR core/serve: Path Resolve error: context deadline exceeded gateway_handler.go:436
<whyrusleeping> Tv`: cool, was trying to write something to parse the subnet notation a.b.c.d/n
<whyrusleeping> and build address filters to fix our excessive dialing issue
<whyrusleeping> s/fix/mend slightly/
therealplato has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
flugsio has quit [Quit: WeeChat 1.2]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
wallet42 has joined #ipfs
Wallacoloo has joined #ipfs
therealplato has quit [Ping timeout: 240 seconds]
www1 has joined #ipfs
www has quit [Ping timeout: 245 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<Tv`> whyrusleeping: don't write that stuff yourself!
anshukla has quit [Remote host closed the connection]
domanic has joined #ipfs
lgierth_ has joined #ipfs
pfraze_ has quit [Ping timeout: 256 seconds]
pfraze has joined #ipfs
lgierth has quit [Ping timeout: 240 seconds]
<whyrusleeping> Tv` oh, that's nice
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<peerchemist> @whyrusleeping; I keep getting this error "Error: context deadline exceeded" when trying to list files on another node or pull file from there
<peerchemist> pinging works
<whyrusleeping> so you can ping the node and you can perform the action just fine on the other node?
<peerchemist> yep
<peerchemist> but "ls" and "cat" fail with "Error: context deadline exceeded"
<whyrusleeping> ls and cat work on the other machine?
<peerchemist> on neither
<whyrusleeping> are you sure the content in question exists?
<peerchemist> yup
<whyrusleeping> and that you have the hash correct?
<peerchemist> also, "ls" should work
<peerchemist> yes, I use the same hash to ping and it works
<whyrusleeping> oh
<whyrusleeping> that's not how it works
<whyrusleeping> the hash you ping is the peers id
pfraze has quit [Remote host closed the connection]
domanic has quit [Ping timeout: 244 seconds]
<peerchemist> and it is different for files?
<whyrusleeping> it's the hash of your public key
<whyrusleeping> ideally catting it should spit out your public key, but we currently special case public keys
<peerchemist> okay, so I need a hash of the file to cat it? how come you use filename in your example?
<peerchemist> or I am totally wrong
<peerchemist> can you cat a file from my node?
<peerchemist> QmSWwyjW9UFiTrk9wwkRZMLAPCzbNqwcRieZf9XVfAxsky/J3IWZtU.jpg
rschulman has quit [Quit: leaving]
wallet42 has quit [Quit: Leaving.]
<Blame> that pulls your file from a gateway
<Blame> and no, it does not seem to be working
<peerchemist> yup
<peerchemist> never mind, shutting down for now
<peerchemist> will try again later
<peerchemist> thx all for help and conversation
notduncansmith has joined #ipfs
<peerchemist> I will def stick around this some more
notduncansmith has quit [Read error: Connection reset by peer]
inconshr_ has joined #ipfs
therealplato has joined #ipfs
inconshreveable has quit [Ping timeout: 265 seconds]
se3000 has quit [Quit: se3000]
peerchemist has quit [Quit: Leaving.]
anshukla has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
timgws has joined #ipfs