<daviddias>
dignifiedquire: doing it now (checking that it passes with version 2 of phantom)
<dignifiedquire>
you are awesome :)
<daviddias>
release new versions of all of the things! :D
cristian_ has joined #ipfs
Looking has joined #ipfs
jaboja has joined #ipfs
<dignifiedquire>
:)
<dignifiedquire>
that's what mondays are for
<ipfsbot>
[js-ipfs] diasdavid deleted greenkeeper-ipfs-blocks-0.1.2 at 9bdff64: https://git.io/vV785
cristian_ has quit [Ping timeout: 244 seconds]
tmg has joined #ipfs
afisk has joined #ipfs
<daviddias>
dignifiedquire: I keep getting the ` Uncaught Error: Bad dictionary `, although I changed on the webpack conf inside the karma conf to replace the browserify-zlib module with our own
Not_ has joined #ipfs
<dignifiedquire>
daviddias: hmm did you check the source code in node_modules is the right one?
<daviddias>
yep, it is the one from ipfs/browserify-zlib
afisk has quit [Ping timeout: 248 seconds]
<daviddias>
dignifiedquire: can you reproduce?
<dignifiedquire>
it's working on my machine, will try on another machine in a sec
afisk has joined #ipfs
nycoliver has quit [Ping timeout: 240 seconds]
<daviddias>
dignifiedquire: any special change in karma.conf?
<dignifiedquire>
not that I know abotu
<dignifiedquire>
it's failing now on my machine where it worked before -.- but the dict tests on browserify-zlib pass
<daviddias>
ah, but I mean, on libp2p-spdy, you had tested it too, right?
jaboja has quit [Ping timeout: 240 seconds]
<dignifiedquire>
I had
<dignifiedquire>
but the tests are not passing anymore
<dignifiedquire>
and i don't understand why :(
<daviddias>
oh :(
<daviddias>
do they fail with 'bad dictionary' ?
<dignifiedquire>
yes
<dignifiedquire>
which is better than "missing dictionary"
afisk has quit [Read error: Connection reset by peer]
Wolf480pl has quit [Ping timeout: 264 seconds]
Wolf480pl has joined #ipfs
<nicolagreco>
dignifiedquire: amazing, I have a write up draft that I will post soon about the conversations that I had with jbenet, so that we can all review it
rschulman has joined #ipfs
gunn_ has quit [Ping timeout: 244 seconds]
gunn has joined #ipfs
jaboja has joined #ipfs
gigq has joined #ipfs
ylp1 has quit [Quit: Leaving.]
bsm117532 has joined #ipfs
Oatmeal has joined #ipfs
<dignifiedquire>
nicolagreco: great
<dignifiedquire>
nicolagreco: looking forward to reading it
Akaibu has quit [Quit: Connection closed for inactivity]
mildred has quit [Quit: Leaving.]
<voxelot_>
nicolagreco: let me know if you need me to takes pics of notebook you were using
cristian_ has joined #ipfs
Boomerang has quit [Ping timeout: 276 seconds]
Boomerang has joined #ipfs
<noffle>
mornin'
voxelot_ has quit [Remote host closed the connection]
Boomerang has quit [Quit: Leaving]
tmg has quit [Ping timeout: 260 seconds]
voxelot_ has joined #ipfs
<dignifiedquire>
noffle: good morning
<dignifiedquire>
richardlitt: are you alive?
afisk has joined #ipfs
pfraze has joined #ipfs
<voxelot_>
morning noffle!
<voxelot_>
mid-sprint in 12 minutes?
<dignifiedquire>
I thought in an hour?
<noffle>
EU time change happened, right?
<dignifiedquire>
yeah
<noffle>
so I think we're back to 4pm UTC
afisk has quit [Ping timeout: 248 seconds]
Looking has joined #ipfs
<dignifiedquire>
oh ok
<haad>
wait
<haad>
we doing the hangout in 5 mins or in 1h 5mins?
<noffle>
5mins, as I understand it
<dignifiedquire>
I hope in an hour
<noffle>
y'all did 4pm UTC last week
<haad>
I'm not at home
<haad>
will not get back in 5mins
<noffle>
and I arrived an hour late, b/c I thought we were doing 5pm UTC
<noffle>
but richardlitt told me we were back to the old time
<haad>
ok. well, I hope it is at 7pm my time (1h), otherwise I won't be there.
afisk has joined #ipfs
<dignifiedquire>
well given that richardlitt is not here I suggest we make it in an hour,
TheNain38 has joined #ipfs
<voxelot_>
in NY we confirmed 12pm EST (6 minutes) but richard is not on so not sure whats happening
<voxelot_>
either time works for me
afisk has quit [Remote host closed the connection]
afisk has joined #ipfs
Encrypt has quit [Ping timeout: 246 seconds]
<noffle>
welp
<richardlitt>
good morning
<richardlitt>
Sorry been having a chaotic morning.
<noffle>
sounds like we'd best do 1h late?
<richardlitt>
It should happen in around 2 minutes.
<richardlitt>
Do you guys want to do it in an hour?
<nicolagreco>
does anyone have the pictures of the notebook we used jbenet, daviddias ?
<noffle>
haad won't be around
<richardlitt>
That would work much better for me, actually. Right now I'm missing lunch with my two sisters.
<daviddias>
nicolagreco: voxelot has it :)
<voxelot_>
nicolagreco: will take pics for you right now :D
jaboja has quit [Ping timeout: 244 seconds]
<richardlitt>
haad: will you be here in an hour?
cristian_ has quit [Ping timeout: 244 seconds]
<dignifiedquire>
I can't right now either but in an hour
cemerick has quit [Ping timeout: 264 seconds]
<richardlitt>
Ok.
<richardlitt>
Let's meet in an hour
<dignifiedquire>
haad said he will be around in an hour
<richardlitt>
Yeah. If both of you will miss it, and if I would prefer it, consider it done.
<richardlitt>
See you all in an hour, then!
afisk has quit [Remote host closed the connection]
null_rad- has quit [Quit: EliteBNC free bnc service - http://elitebnc.org - be a part of the Elite!]
onla has joined #ipfs
fiatjaf_ has joined #ipfs
onla has quit [Ping timeout: 250 seconds]
onla has joined #ipfs
<voxelot_>
whyrusleeping: i put study go-lang on my to-do
<richardlitt>
thanks all!
Wolf480pl has joined #ipfs
holst has joined #ipfs
<holst>
WOW much people. happy to see not only me are exicted about this stuff ;-)
<holst>
I wonder if ipfs supports encryption or if I have to start a virtual machine, open a luks encrypted partition and serve the stuff over to windows?
<holst>
I want to store the data on a portable USB disk
<akkad>
anyone ever use ipfs to cache large things like docker images?
holst has quit [Read error: Connection reset by peer]
<whyrusleeping>
akkad: yeah, i had a quick hack a while ago to move docker images
<whyrusleeping>
basically docker export into ipfs
holst has joined #ipfs
<whyrusleeping>
then import from ipfs into docker on another machine
<lgierth>
yeah docker export (for containers) and save (for images)
<lgierth>
and you can probably even use `ipfs tar` for it
ygrek has joined #ipfs
<whyrusleeping>
yeap!
<whyrusleeping>
we were blocked on that a while ago because ipfs couldnt handle symlinks
* holst
google ipfs irc chat logs
<whyrusleeping>
but now it can
<whyrusleeping>
holst: theyre over at botbot
<whyrusleeping>
theres a link in the topic
<holst>
ah thanks. I just wanted to check if I where seeing a tail of an answer to my Q ;)
<nicolagreco>
Oh no I wanted to join the sprint this week (but I am at the WWW conference and I had an event clashing)
Encrypt has quit [Quit: Eating time!]
null_radix has joined #ipfs
zorglub27 has quit [Ping timeout: 244 seconds]
bsm117532 has quit [Ping timeout: 250 seconds]
matoro has quit [Ping timeout: 244 seconds]
ola has joined #ipfs
cristian_ has joined #ipfs
<jbenet>
daviddias: im on it (re unblocking you for core-api spec and registry-mirror things)
<jbenet>
dignifiedquire: awesome re ipld resolver!
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
palkeo has joined #ipfs
<dignifiedquire>
jbenet: thanks, did you try out ipld-cli yet?
<nicolagreco>
dignifiedquire, jbenet I posted the notes I got from the conversation we had, I hope I didn't make it confusing with the terms I picked https://github.com/ipfs/specs/issues/91
<dignifiedquire>
thanks nicolagreco will read them tomorrow
<nicolagreco>
dignifiedquire: the command line tool is exactly what I wanted!
<dignifiedquire>
nicolagreco: happy to hear that
<nicolagreco>
even the colouring, you made it so great ;)
<nicolagreco>
I just think we need to do some progress on the ipld spec itself and I think it will be great!
neurosis12 has joined #ipfs
neurosis12 has joined #ipfs
Akaibu has quit [Quit: Connection closed for inactivity]
zorglub27 has joined #ipfs
jaboja has quit [Ping timeout: 244 seconds]
zeroish has joined #ipfs
zorglub27 has quit [Quit: zorglub27]
Guest92941 is now known as Neozonz
Neozonz has quit [Changing host]
Neozonz has joined #ipfs
jaboja has joined #ipfs
Neozonz|Disc has joined #ipfs
Neozonz|Disc is now known as Guest9109
computerfreak has joined #ipfs
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
Neozonz has quit [Ping timeout: 244 seconds]
matoro has quit [Ping timeout: 252 seconds]
Guest9109 is now known as Neozonz
Neozonz has quit [Changing host]
Neozonz has joined #ipfs
supertyler has joined #ipfs
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #321: Update react-router-bootstrap to version 0.21.0
Akaibu has joined #ipfs
<supertyler>
Quick question - Where are the core IPFS team physically located? Anyone in London?
<lgierth>
closest to london would be berlin
cryptix has quit [Ping timeout: 244 seconds]
<supertyler>
thanks - who is the contact there?
<lgierth>
that'd be haad and me
cristian_ has quit [Ping timeout: 244 seconds]
matoro has joined #ipfs
cristian_ has joined #ipfs
Neozonz has quit [Read error: Connection reset by peer]
Neozonz has joined #ipfs
Neozonz is now known as Guest75083
rendar has quit [Ping timeout: 246 seconds]
jfis has joined #ipfs
montagsoup has joined #ipfs
Encrypt has joined #ipfs
rendar has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping created fix/libp2p-listen-hang (+1 new commit): https://git.io/vVdnj
<ipfsbot>
go-ipfs/fix/libp2p-listen-hang bdc5456 Jeromy: update libp2p dep to fix hanging listeners problem...
Oatmeal has quit [Ping timeout: 250 seconds]
fiatjaf_ has quit [Remote host closed the connection]
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #2548: update libp2p dep to fix hanging listeners problem (master...fix/libp2p-listen-hang) https://git.io/vVdWP
cristian_ has quit [Ping timeout: 244 seconds]
<montagsoup>
anyone know about how fast ipns is? I dont need anything exact, just a general feel for the speed
Temper has quit []
pfraze has quit [Remote host closed the connection]
Guest75083 has quit [Ping timeout: 260 seconds]
Oatmeal has joined #ipfs
<ipfsbot>
[go-ipfs] jbenet deleted fix/libp2p-listen-hang at bdc5456: https://git.io/vVdRc
<whyrusleeping>
montagsoup: i get resolutions in around 3 seconds normally
<whyrusleeping>
its going to be faster, we havent put too much effort into perf there
<whyrusleeping>
but its not atrocious for what it does
erictapen has joined #ipfs
cemerick has joined #ipfs
<richardlitt>
anyone have any idea why illustrator might not accurately show an SVG file?
<richardlitt>
Chrome and Illustrator are showing different things for my .svg
<richardlitt>
dignifiedquire?
<dignifiedquire>
hmm corruption on the wrapper maybe
nonaTure has quit [Ping timeout: 250 seconds]
<montagsoup>
whyrusleeping: I'm just considering something that needs to run really fast and I want to figure out where I'll be able to use ipns
<montagsoup>
3 seconds is faster than I was hoping so I should be able to put it is some of the more speed critical places
<holst>
cat went much faster; 0.282 0.100 0.136 (r/u/s)
pfraze has quit [Remote host closed the connection]
<holst>
should probably avg. over multiple runs
<holst>
good thing with the leveldb is that it compresses
<holst>
you can add a sparse 100GB file w/o breaking a sweat
<montagsoup>
I'm not really worried about to speed it takes on a single host. What I'm worried about is how long it takes for other nodes on the network to pick up the update
<montagsoup>
nice to know it adds stuff pretty fast though
<chungy>
I'm in Seattle and see connections from India, Singapore, and Australia... one each. (and I think I have more European connections than American ones :P)
<chungy>
Just kind of cool :)
<holst>
the magic is not in node to node throughput
<holst>
but in the large scale orcestration
<chungy>
yea
sahib has joined #ipfs
<chungy>
I've done little intra-LAN tests and love how fast it is.
<chungy>
Not quite as good as CIFS or NFS, but still decent
<holst>
5min to add a 100GB file
<holst>
on a single core, but i assume the software is not multithreaded yet?
<holst>
the chunk write/check/ could probably be trivial to dispatch over multiple threads
<holst>
sry. thats a _10_ gb file
<chungy>
fwiw, just saw the pastebin, you can use "truncate -s 100M 100MB" to create the same file.
<holst>
pretty sad if you want to add 7500 GB of data.
<holst>
running on SSD disk so io bw is not the limiting factor (the single cpu core might be)
<montagsoup>
also to be specific, I'm trying to find out the speed of just ipNs. I'm looking into how i might be able to use it for sigining of really fast (real time) updates
<lgierth>
yeah `ipfs add` is cpu-intensive because of all the hashing
<chungy>
the 100MB add takes 22s for me... I have a pretty old system, granted
<chungy>
lgierth: maybe it could be smarter with sparse files?
cryptix has quit [Ping timeout: 244 seconds]
<holst>
sys 0m32.808s
<holst>
thats presumably waiting-for-io
<lgierth>
it's also completely serial at the moment
<lgierth>
that means read, chunk, hash, write, repeat
<holst>
lgierth: I assumed so. no real problem its "low hanging fruits" later for the optimization phase :)
<lgierth>
yeah :)
<lgierth>
there's an old parallelism-for-add PR in go-ipfs.git
<holst>
the good thing is that "cat" has virtually no overhead for the sparse10g file
<holst>
but the hasing seems to be much slower than the one implemented in for instance openssl
<holst>
openssl sha256 runs through the same file in less than 1:30
<lgierth>
ipfs doesn't hash the file/data itself
<lgierth>
it chunks it, wraps it in a datastructure, links the chunks
Guest75083 has joined #ipfs
<chungy>
Is there a plan to stop advertising non-routable addresses for nodes? All the private IPv4 and IPv6 blocks.
<chungy>
I imagine it's a non-issue for home users, but it could be a security breach in corps...
TomasBond has joined #ipfs
<holst>
what is a non-routable address?
<TomasBond>
who is neptune.i.ipfs
<TomasBond>
why i have so much traffic to that specific node
<lgierth>
TomasBond: that's one of the bootstrap nodes
<holst>
lgierth: does the node communicate with the leveldb compressed data or does it use the unpacked data?
<lgierth>
holst: he means address ranges that are not routable on the internet, e.g. 172.17.0.0/16
<lgierth>
holst: not sure what you mean
<chungy>
holst: In IPv6, anything that's not 2000::/8. In IPv4, you have 10/8, 172.16/12, 192.168/16, 169.254/16, and a few others
<holst>
I just added 10GiB of zeros
<holst>
would it send 10GB or the compressed representation of that to any replicas?
<chungy>
I think in IPv6, it's strictly just f000::/8 that's a private/link-local address, but currently only 2000::/8 is assigned for public internet
cemerick has quit [Ping timeout: 244 seconds]
<lgierth>
chungy: i think you mean fc00::/7 re: private
<chungy>
when you look at node details, it gives information about *all* interfaces the node is bound to
<lgierth>
and fe80::/16 re: link-local
jaboja has quit [Ping timeout: 250 seconds]
Senji has quit [Ping timeout: 244 seconds]
<chungy>
*nods* Not intending to give a full breakdown of non-routable addresses. Suffice to say that f000::/4 isn't routable.
Guest75083 is now known as Neozonz
Neozonz has quit [Changing host]
Neozonz has joined #ipfs
<lgierth>
aah /4
<lgierth>
word
Neozonz has quit [Quit: Leaving]
<chungy>
It's a little bit tricky. You want other nodes in the same subnet to be aware of the local addresses.
Neozonz has joined #ipfs
<chungy>
Including the f000::/4 addresses in v6, and all the various subnets a node might be part of on v4.
<montagsoup>
cjdns uses one of the private blocks, so it might be best to keep it to stay compatable
<holst>
I think I just answered my own question
<holst>
the 10G file was stored as the "IPFS object(s)"
Senji has joined #ipfs
<holst>
s/the/three/
<chungy>
holst: heh, yeah, it's going to be repeating the same block multiple times
<holst>
funny thin is it partitioned the files in two different ways
<holst>
the 100MB file and the 10GB file
Looking has quit [Ping timeout: 276 seconds]
<chungy>
I made a 4.5PB tree on 0.3.11 once. Maybe I should repeat that on 0.4.0
supertyler has quit [Ping timeout: 250 seconds]
<lgierth>
chungy: yes please
<holst>
100MB file was splitted into 3 objects; where the first and second object are the same
neurosis12 has quit [Remote host closed the connection]
<holst>
but the 10GB file was split into two objects
<chungy>
lgierth: haha :)
taw00 has quit [Ping timeout: 264 seconds]
<chungy>
It'll be cool to do something like jigdo... where, for example, a Debian *.iso file can be broken up into blocks identical to individual files, so it'd be deduplicated with a regular tree of packages.
<lgierth>
yeah -- the default chunker just splits at 256k, but there's the rabin fingerprinting chunker which is pretty smart
<lgierth>
ipfs add -s rabin
<lgierth>
hehe i guess that's all just /dev/zero?
<chungy>
0xaa bytes, actually
afisk has quit [Remote host closed the connection]
Encrypt has quit [Quit: Quitte]
rmi7 has quit [Ping timeout: 248 seconds]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
ruby32 has quit []
joshbuddy has joined #ipfs
cryptix has joined #ipfs
matoro has quit [Ping timeout: 244 seconds]
cryptix has quit [Client Quit]
<peterix>
so... let's say I want to publish a pile of json files that changes very often. how do I tell the client that is supposed to use these that the collection changed?
devbug has joined #ipfs
<peterix>
right now, I get a hash... and it needs to somehow get to 200000 machines somewhere out there
<noffle>
peterix: IPNS is one way of retaining a mutable pointer to an IPFS address (e.g. ipfs name publish /ipfs/Qmfoobar)
<noffle>
those machines could know of that /ipns/... address ahead of time and poll it on intervals
<noffle>
more realtime solutions are Coming Soon (lots in the ipfs/notes repo issues)
<peterix>
so. integrating ipfs with desktop applications. I assume using the current implementation as is and just running separate processes would work
<ipfsbot>
[go-ipfs] lgierth created datastore-config (+1 new commit): https://git.io/vVdho
<ipfsbot>
go-ipfs/datastore-config aae44f8 Lars Gierth: repo: properly init Datastore config, and leave it be...
<ipfsbot>
[go-ipfs] lgierth force-pushed datastore-config from aae44f8 to 25e8e7a: https://git.io/vVdhQ
<ipfsbot>
go-ipfs/datastore-config 25e8e7a Lars Gierth: repo: properly init Datastore config, and leave it be...
<ipfsbot>
[go-ipfs] lgierth opened pull request #2549: repo: properly init Datastore config, and leave it be (master...datastore-config) https://git.io/vVdhN
Oatmeal has quit [Ping timeout: 250 seconds]
TheNain38 has quit [Quit: I'm going away]
<ipfsbot>
[go-ipfs] lgierth force-pushed datastore-config from 25e8e7a to a79907b: https://git.io/vVdhQ
<ipfsbot>
go-ipfs/datastore-config a79907b Lars Gierth: repo: properly init Datastore config, and leave it be...
devbug has quit [Quit: Lances at dawn!]
kvda has joined #ipfs
kaiza has quit [Quit: o7]
Seka has quit [Read error: Connection reset by peer]
tmg has joined #ipfs
pfraze has quit [Remote host closed the connection]