<jbenet>
freedaemon: i owe you and ryan some email-- sorry been slammed with things :)
patcon has quit [Ping timeout: 276 seconds]
<void>
jbenet, whyrusleeping: What are the requirements to create a branch in the repo?
<jbenet>
void: you mean to have push access to the main repo?
<jbenet>
void: because you can fork + create PRs as much as you want.
<void>
jbenet: examples
<void>
okay, I will fork then.
<jbenet>
void: you can gain push access to the main repo after a few merged PRs so we all make sure we're good on the flow.
<void>
great, thanks.
freedaemon has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
tso has left #ipfs [#ipfs]
domanic has quit [Ping timeout: 252 seconds]
riveter has quit [Ping timeout: 272 seconds]
cjb has quit [Ping timeout: 256 seconds]
riveter has joined #ipfs
tilgovi has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
fleeky has quit [Ping timeout: 264 seconds]
hellertime has joined #ipfs
fleeky has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
void has left #ipfs [#ipfs]
<whyrusleeping>
jbenet: may have found a bug in multiaddr net
Wallacoloo has joined #ipfs
<whyrusleeping>
oh wait, nvm
<whyrusleeping>
mistyped something
<whyrusleeping>
as usual
<zignig>
whyrusleeping: Realli yor tiping is really good most of the tome.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<zignig>
:P
<whyrusleeping>
lol
<whyrusleeping>
you makin fun of me over there?
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to fix/addr-filter: http://git.io/vtHkf
<ipfsbot>
go-ipfs/fix/addr-filter 7cc73f7 Jeromy: add command to manipulate address filters and a sharness test for them...
<zignig>
no.... :D
<zignig>
how goes dev in general ? onwards and upwards ?
<whyrusleeping>
onwards and upwards!
<whyrusleeping>
have a lot of perf improvements in the works
<whyrusleeping>
one that more than doubles add speed for large files
<whyrusleeping>
and a protocol change that should reduce latency throughout the entire system by hopefull half
<zignig>
great ! , getting files from behind a nat can be painful.
<whyrusleeping>
(that latter one is whats known as a "bugfix")
<lgierth>
coolio
<zignig>
sound good, getting that latency down is important, also dealing with NAT crappyness.
<zignig>
I was having a look at logs on full debug (wow) , there still seems to be a large amount of failed dialing.
<zignig>
I can also confirm that the multicast advertisment works well , a 15 node ipfs cluster found all of it's local friends.
<zignig>
can you make it so the dialer checks local ( as in mdns ) nodes first for blocks ?
<whyrusleeping>
zignig: it does so without really trying
<whyrusleeping>
currently ipfs is set up to always be connected to mdns nodes
<whyrusleeping>
and we broadcast our 'wants' to all connected nodes before we even start searching the dht for providers
<whyrusleeping>
so if nodes on your LAN have blocks, youre going to get them pretty immediately
<spikebike>
nice
<spikebike>
I like having it always work.. but fast if you have a direct connection
<lgierth>
something like bittorrent's retracker would be neat, essentially making others' "wants" your own too
<lgierth>
kind of a cache
notduncansmith has joined #ipfs
<whyrusleeping>
lgierth: yeah, we've talked about that
<whyrusleeping>
it would be an off-by-default feature though
notduncansmith has quit [Read error: Connection reset by peer]
<lgierth>
yeah
<lgierth>
ouch. according to the metrics, uranus has -100 peers
<whyrusleeping>
>.>
<whyrusleeping>
those must be lingering connections
<lgierth>
that get killed multiple times?
<whyrusleeping>
yeah
<whyrusleeping>
i dont think peers are removed from that list when they disconnect
<whyrusleeping>
ipfs diag net normally only shows around 70
Wallacoloo has left #ipfs [#ipfs]
<lgierth>
i mean the prometheus metrics merged today, the gateways already run that
<whyrusleeping>
i'd love it if our network were > 100 nodes
<lgierth>
it's Inc() onConnect, and Dec() onDisconnect
<lgierth>
somehow that got out of balance on uranus
<pjz>
once I 'ipfs ađd' a file, I can nuke it from my drive, right? it's been turned into pinned blocks in my ipfs storẻ
<zignig>
whyrusleeping: with the config template on the init and auto port select.
<zignig>
whyrusleeping: alstralboot could fire up as many nodes as fits.
<zignig>
*astralboot
<whyrusleeping>
zignig: i beleive so
<zignig>
ipfs init --config=<file> and a zero port on the swarm address, WIN!
<lgierth>
same of mercury now
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ruby32 has joined #ipfs
pfraze has quit [Remote host closed the connection]
ruby32 has quit [Ping timeout: 256 seconds]
dread-alexandria has joined #ipfs
<jbenet>
whyrusleeping: re "currently ipfs is set up to always be connected to mdns nodes" maybe `ipfs init --interactive` could ask the user what to do
<jbenet>
lgierth: "uranus has -100 peers" damn. :( that means a) our notifications are firing more than often (which may be unavoidable). and b) maybe we should "record a change" upon the notification, but ask the swarm how big it is at that time.
<jbenet>
currently the network doesn't expose the swarm / host / number of conns. but maybe we can get it.
<jbenet>
we could expose a "Peers()" or something
<lgierth>
jbenet: right that'd be more robust -- mercury was below 0 briefly too
<lgierth>
anyhow, all the counts are wrong
<lgierth>
i'm about to go to bed so i'll fix that tomorrow
warptangent has quit [Remote host closed the connection]
warptangent has joined #ipfs
notduncansmith has joined #ipfs
<jbenet>
lgierth: sounds good
O83110 has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
tilgovi has quit [Ping timeout: 252 seconds]
void has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
dread-alexandria has quit [Quit: dread-alexandria]
hellertime has quit [Quit: Leaving.]
<Leer10>
hey does anyone know of any good ipfs sites I can link to for the side of my ipfs site?
<Leer10>
trying to get something akin to a blogroll
<whyrusleeping>
Leer10: i dont know of many yet, i havent had time to make my own yet :/
<Leer10>
do you know of anything else that ipfs can be used for besides a place to put sites and files in?
Wallacoloo has joined #ipfs
<whyrusleeping>
Leer10: what do you use the internet for?
<void>
Leer10: Everything can be a file. So can hardware. Or processes.
<Leer10>
so you're saying that ipfs can be expanded to access computers (in a way)
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<void>
yes
<reit>
there's not currently a messaging layer if that's what you're asking, but it should be possible to add one in the future
<whyrusleeping>
you can write your own messaging layer, we provide all the primitive currently
<whyrusleeping>
and there will be nice tooling around it in the future
<prosodyContext>
jbenet I was reading them chat about it in bits a few times, something about Amandine one of the cofounders they controlled the car somehow. I'm sure it was just a starter ;)
<prosodyContext>
Has anyone spelled IPFSAI yet? O:)
<Leer10>
ooh nice
<Leer10>
I am also aware of the whole decentralized thing where the network can split and some things can still function
<Leer10>
but does that also mean that if there's a bunch of copies of the same file does that make access of that file faster (as in speed) or does it merely make it quicker to access (as in ping)
<void>
Leer10: Latency may be reduced if other peers are physically closer to you, due to the architecture of the (bitswap/bittorrent) protocol you can also receive your data from multiple nodes at the same time.
<Leer10>
Cool
<Leer10>
now are non-seekable live streams still possible?
M-_mis has quit [Remote host closed the connection]
MatrixBridge2401 has quit [Read error: Connection reset by peer]
MatrixBridge has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
MatrixBridge has joined #ipfs
dread-alexandria has quit [Quit: dread-alexandria]
<jbenet>
Leer10: i love your guide. in particular the hashnames ;)
<Leer10>
haha thanks
<Leer10>
I'll be adding more stuff like how to add posts and embedding pics but what is on there is enough to get someone off the ground
<Leer10>
probably tomorrow though cause it's late
atomotic has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<temet>
I'm creating accounting wrapping scripts, would you like me to contribute them as a set of tools to manage more of the metadata regarding files added and such?
kbala has quit [Quit: Connection closed for inactivity]
<jbenet>
temet please do! Not sure go-Ipfs is the best repo but we should collect them somewhere
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<temet>
For now, i'll put them up on a personal repo as it polishes up.
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Tv` has quit [Quit: Connection closed for inactivity]
equim has quit [Ping timeout: 264 seconds]
equim has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
ruby32 has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
bedeho2 is now known as bedeho
domanic has joined #ipfs
ruby32 has quit [Ping timeout: 256 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]
notduncansmith has quit [Read error: Connection reset by peer]
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]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
therealplato has quit [Ping timeout: 248 seconds]
therealplato has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
atomotic has joined #ipfs
<cryptix>
<cryptix>
ohai :)
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
cjb has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
www has joined #ipfs
<temet>
hmm, interesting question, can i create a "directory" block using a lower level command other than ipfs add -w or ipfs add .
<cryptix>
temet: yup. try 'ipfs object new'
<temet>
thanks
<cryptix>
you want to use the unixfs-dir template (see --help on object new) if you want it to be a filesystem dir
<cryptix>
also see 'ipfs object patch'
<temet>
hahah spitting the contents of an object just totally messed my terminal. Fun stuff.
<temet>
alright thanks
<temet>
then how do i link a data object underneath it
<temet>
i keep mixing up pinning with other functions.
* temet
isn't grokking it very well.
se3000 has quit [Ping timeout: 272 seconds]
<cryptix>
temet: yep - leav pinning out of it
<cryptix>
pinning is only related to what your local repo holds onto when collecting garbage
<temet>
then how do you link an object to the "directory"
<cryptix>
have you seen the help of the object patch command?
notduncansmith has joined #ipfs
<temet>
yeah
<temet>
add-link
<temet>
right
<temet>
okay thanks
notduncansmith has quit [Read error: Connection reset by peer]
<cryptix>
;)
<temet>
Just rememberd
<temet>
:3
<temet>
I am RTFM i swear
<cryptix>
its all right :) just wanted to make sure you saw that
<temet>
yeah, it's a little difficult trying to hold all the command options in my head at once.. i'll end up printing the command page soon
<cryptix>
with enough repition it will make 'click' at some point.. :) just try to seperate high and low level cleanly (and dont mix up pinning with it.. :)
domanic has quit [Ping timeout: 246 seconds]
* cryptix
will go outside now - l8ters
<temet>
bai
therealplato has quit [Ping timeout: 264 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<lgierth>
which i guess is consistent with us updating the metric onConnect/onDisconnect
<jbenet>
lgierth: the former type (Network) is a cast for the Swarm to make it conform to a different API without name clashes or exapanding the interface
<jbenet>
PeerHost.Network().Conns() should be right, too
<lgierth>
cool
<jbenet>
lgierth the comment on "Peerstore()" is wrong: "Peers returns the connected peers"
tilgovi has quit [Ping timeout: 252 seconds]
<lgierth>
i'll update that then
<jbenet>
lgierth it shoule be "Peerstore returns the associated peerstore, which tracks peers known."
<jbenet>
or something
<pjz>
known peers
tilgovi has joined #ipfs
<spikebike>
very strange. I was tinking with ipfs and ipv6. I broke ipv6 for awhile, then fixed it, and the ipfs daemon did list some ipv6 peers but I could add new ones.
<spikebike>
ipfs swarm connect would say success, but ipfs swarm peers wouldn't list the new peer
<spikebike>
telnet -6 <remote IP> 4001 would connect
<spikebike>
restarting ipfs and hitting uparrow to readd the peer workded
<spikebike>
very strange. I was tinkering with ipfs and ipv6. I broke ipv6 for awhile, then fixed it, and the ipfs daemon did list some ipv6 peers but I could not add new ones.
Tristitia has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
Blame has quit [Remote host closed the connection]
Blame has joined #ipfs
<daviddias>
jbenet: search for 'Plan-B-Space-2.local'
e-lima has quit [Ping timeout: 248 seconds]
<whyrusleeping>
also wrote some nodejs, it was awful
<whyrusleeping>
sprint checkin: rebased go-datastore batching PR, and working on locking around GC for my GC pr
<jbenet>
lgierth does arcial == github.com/arceliar?
<lgierth>
yeah
<jbenet>
lgierth i'm going to cc in response to note, cool!
<temet>
Ah, this is a bit mind bending. So now that i'm using the patch command, is there a way to change the "filename" of the links stored underneath a unixfs-dir object
<temet>
?
<whyrusleeping>
temet: like a rename?
<temet>
yes
<whyrusleeping>
it does not appear so... we should add one
<temet>
or rather, how do i creat e a file (with a name attached) then adding it to unixfs-dir node so that i can append files to the directory
<whyrusleeping>
right now you can do a remove-link followed by an add link
<temet>
create*
<temet>
i have an index.html.
<temet>
But now I want to add a .js file
<whyrusleeping>
so ipfs add index.html -> INDEX_HASH
<lgierth>
the importpath is not a url and doesn't behave like one
<lgierth>
it's just a filesystem path, and knows to git clone when it starts with 'github.com'
<whyrusleeping>
lgierth: go get's type inference is quite the crazy mess of code
<lgierth>
can imagine
<kbala>
hey whyrusleeping what are peerstores for?
<kbala>
and how would i remove a peer from a netsim network
<whyrusleeping>
kbala: peerstores are for holding all information about the peers we know
<whyrusleeping>
like their addresses and their public keys
<lgierth>
that's a superset of the swarm peers (the ones we're connect to) btw
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<kbala>
ah ok thanks
<whyrusleeping>
kbala: also, removing a peer from netsim, i'll have to take a look
<kbala>
so does each peer have a peerstore with the info of other peers it knows of? or is there one peerstore for the network that has information about all the peers in the network
<whyrusleeping>
each peer has a peerstore
<whyrusleeping>
thats part of go-ipfs
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed dev0.4.0 from 0e5faec to 617714a: http://git.io/vLnex
<ipfsbot>
go-ipfs/dev0.4.0 79fb4f9 Tommi Virtanen: pin: Rewrite to store pins in IPFS objects...
<ipfsbot>
go-ipfs/dev0.4.0 42b9e53 Tommi Virtanen: pin: Future-proof against refcount marshaled size changes
<ipfsbot>
go-ipfs/dev0.4.0 86776db Jeromy: bump fsrepo version to 3
chrisr_ has quit [Ping timeout: 252 seconds]
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed proto-fix from 018db49 to 956e398: http://git.io/vtNzo
<ipfsbot>
go-ipfs/proto-fix c42b501 Juan Batiz-Benet: remove msgio double wrap...
notduncansmith has quit [Read error: Connection reset by peer]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
pfraze has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<kbala>
whyrusleeping: sorry one more question, whats assumed about the peerstore being passed into the HostOption method? if i want the host of a peer to create a bitswapnetwork instance for it, should i create a peerstore with all the addresses of the other peers in the network?