<noffle>
whyrusleeping: daviddias: heya! \o I wanted to get your thoughts on resolving a node's own public key (i.e. /ipnfs/<my-pubkey>). right now, in go-ipfs, this is not preloaded into the local dht store, so what commonly happens is either a) you need the network (and for your pubkey to exist in the network dht), or b) you're offline and do have your own pubkey in the local dht, but its expired (TTL)
tmg has quit [Quit: Reconnecting]
<noffle>
what if we preloaded it into the local dht when the daemon starts? this seems like a key a node should always be able to instantly resolve
dignifiedquire has quit [Quit: Connection closed for inactivity]
gaboose has quit [Remote host closed the connection]
<whyrusleeping>
noffle: it is loaded into the local dht store
<ipfsbot>
[go-ipfs] rht closed pull request #2246: More robust version checking (master...smarter-launchd-install) https://git.io/vzKMm
<noffle>
whyrusleeping: can you give me a pointer as to where? I didn't see that. and if we are then there's something still wrong: I can't resolve my own key for a fresh node that's offline
reit has quit [Quit: Leaving]
alterego has joined #ipfs
captain_morgan has quit [Remote host closed the connection]
HostFat has quit [Quit: Leaving]
<lightcastshadows>
tmg: No, not anymore. It returns fine now. What was the issue?
voxelot_ has joined #ipfs
reincite has quit [Remote host closed the connection]
<voxelot_>
whyrusleeping: o/
<whyrusleeping>
voxelot_: heyo
<whyrusleeping>
noffle: its saved in the datastore
<whyrusleeping>
and the dht pulls from the datastore
<noffle>
whyrusleeping: does this happen on init? daemon start?
<whyrusleeping>
the reason youre seeing the resolve hit the network is because we don't trust our local value to be the most up to date
<aaaaaaaaa____>
another pinning question: pinning 350gb directory melts down (high cpu/ram); so i exported a list of 40,000 hashes and with bash script try to pin them one by one, but this also grinds to a halt after a few hundred, with no message in daemon log. don't know where to go from here, but curious if it is a known issue?
<voxelot_>
that says that we marshal the private key, which means creating a protobuf of the ANS.1 private key stream right?
<whyrusleeping>
aaaaaaaaa____: mind filing an issue?
<aaaaaaaaa____>
sure, will do
<whyrusleeping>
voxelot_: uhm, yes
<whyrusleeping>
the protobuf has the key type, and the data in it
<voxelot_>
whyrusleeping: yup, and then the config file stores a base64 encoded protobuf
<voxelot_>
which is why i can't use it haha
<noffle>
whyrusleeping: humour me a bit more? why have the local cache then if we don't trust it to be up to date?
<whyrusleeping>
voxelot_: yeah... just unwrap the protobuf first
<whyrusleeping>
it should work for you
<voxelot_>
in javascraps
<voxelot_>
there is a module for it, not sure if david has a better way though
<whyrusleeping>
voxelot_: thats okay, javascraps has protobuf
<voxelot_>
okay cool, i'll use that module then
<whyrusleeping>
noffle: we have the value locally because otherwise how would we keep it around?
<whyrusleeping>
the 'cache' is not stored on disk
<whyrusleeping>
just the value
simonv3 has quit [Quit: Connection closed for inactivity]
<whyrusleeping>
so we have no idea just from the disk that we 'cached value X for Y minutes'
<noffle>
whyrusleeping: if "we hit the network to resolve our own key b/c we don't trust the local value to be the most up to date" (yours from above), then why store it in the local cache at all? this seems contradictory? trying to grok this.
<noffle>
isn't the cache only used to resolve keys?
<whyrusleeping>
the network is really just a bunch of other nodes with values stored on disk
<whyrusleeping>
if nobody stores these values then we have nothing
<noffle>
I understand that. I'm trying to reconcile your statement above though
<noffle>
are we writing it to our local dht just so it replicates to the rest of the network?
<noffle>
(as opposed to, say, for the node's own use)
notdaniel has joined #ipfs
acetylcarbazole has joined #ipfs
computerfreak has quit [Quit: Leaving.]
<whyrusleeping>
our local value is one of the replicated dht values
<whyrusleeping>
its not the canonical source of truth
<whyrusleeping>
just one of the values that exists related to that key
<noffle>
whyrusleeping: in that we don't differentiate between what we wrote during ipfs init vs what came from the network?
<whyrusleeping>
yes
<noffle>
neither is more "truthy"?
<noffle>
col
<whyrusleeping>
correct
<noffle>
cool, too
<noffle>
that helps a lot -- thank you
<whyrusleeping>
which seems wonky to me really
<whyrusleeping>
but its hard to do it any other way
simonv3 has joined #ipfs
<whyrusleeping>
since it needs to be set during init
<whyrusleeping>
and we arent on the network during init
captain_morgan has joined #ipfs
ygrek_ has quit [Ping timeout: 245 seconds]
<tmg>
whyrusleeping: is there a way to force go-ipfs to store blocks using their multihash?
<tmg>
whyrusleeping: or better! how do I get from "12200356df0accadd21bd74695b4ddc700de5f590674a1a188e1a7db3bc1472f998d.data" back to the multihash?
<whyrusleeping>
tmg: decode that name as hex
<whyrusleeping>
and then cast it as a multiaddr
<whyrusleeping>
we cant use multihash base58 for a couple reasons
<whyrusleeping>
one: windows and osx have case insensetive filesystems
<whyrusleeping>
two: base58 encoding is reallllllly slow
<tmg>
(something about some filesystems equating cases)
<lgierth>
interesting
<lgierth>
whyrusleeping: is that an fs-repo implementation detail?
<tmg>
lgierth: the details are in datastore, I think
<ipfsbot>
[go-ipfs] rht closed pull request #2102: Close errnotifier chan on err (master...fix/logwriter-stall) https://git.io/v0Fym
<ipfsbot>
[go-ipfs] rht opened pull request #2279: Close errnotifier chan on err (master...fix/logwriter-stall) https://git.io/vgUiW
simpbrain has quit [Ping timeout: 250 seconds]
ygrek_ has joined #ipfs
<tmg>
whyrusleeping: why call DAG vertices "nodes"?
<tmg>
sort of overloads the name space with IpfsNodes too
pfraze has quit [Remote host closed the connection]
reit has joined #ipfs
simpbrain has joined #ipfs
The_8472 has quit [Ping timeout: 240 seconds]
<Kubuxu>
patagonicus: Yes it does, the CLI just not handle it correctly. It doesn't happen on ipfs.io as it uses /ipfs/ entry in dnslink, you probably use /ipns/.
The_8472 has joined #ipfs
IlanGodik has joined #ipfs
aquilax has joined #ipfs
<ipfsbot>
[go-ipfs] lgierth created feat/gx-docker (+1 new commit): https://git.io/vgUMx
<ipfsbot>
go-ipfs/feat/gx-docker 032aedd Lars Gierth: WIP...
hellertime has quit [Ping timeout: 240 seconds]
The_8472 has quit [Ping timeout: 264 seconds]
Not_ has quit [Ping timeout: 240 seconds]
The_8472 has joined #ipfs
The_8472 has quit [Ping timeout: 240 seconds]
pfraze has joined #ipfs
The_8472 has joined #ipfs
aquilax has quit [Ping timeout: 248 seconds]
<whyrusleeping>
tmg: vertices might be a good name
<whyrusleeping>
i agree that node is way too overloaded
tmg has quit [Ping timeout: 260 seconds]
The_8472 has quit [Ping timeout: 240 seconds]
pfraze has quit [Remote host closed the connection]
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to feat/active-cmds: https://git.io/vgUSp
<ipfsbot>
go-ipfs/feat/active-cmds 4753e6c Jeromy: sort options and add id to verbose output...
The_8472 has joined #ipfs
The_8472 has quit [Ping timeout: 264 seconds]
simonv3 has quit [Quit: Connection closed for inactivity]
Tv` has quit [Quit: Connection closed for inactivity]
The_8472 has joined #ipfs
ygrek_ has quit [Ping timeout: 264 seconds]
The_8472 has quit [Ping timeout: 264 seconds]
The_8472 has joined #ipfs
Guest17506 has quit [Ping timeout: 276 seconds]
notdaniel has quit [Ping timeout: 260 seconds]
The_8472 has quit [Ping timeout: 264 seconds]
zorglub27 has joined #ipfs
The_8472 has joined #ipfs
pepesza_ has quit [Ping timeout: 276 seconds]
pepesza_ has joined #ipfs
jhulten_ has quit [Ping timeout: 264 seconds]
cow_2001 has joined #ipfs
The_8472 has quit [Ping timeout: 240 seconds]
ulrichard has joined #ipfs
zorglub27 has quit [Ping timeout: 272 seconds]
pfraze has joined #ipfs
mildred has joined #ipfs
The_8472 has joined #ipfs
pepesza_ has quit [Excess Flood]
The_8472 has quit [Ping timeout: 264 seconds]
The_8472 has joined #ipfs
dignifiedquire has joined #ipfs
ylp has joined #ipfs
The_8472 has quit [Ping timeout: 240 seconds]
jhulten_ has joined #ipfs
IlanGodik has quit [Quit: Connection closed for inactivity]
acetylcarbazole has quit [Read error: Connection reset by peer]
The_8472 has joined #ipfs
pepesza_ has joined #ipfs
blepharosyndesmi has joined #ipfs
pepesza_ has quit [Excess Flood]
pepesza_ has joined #ipfs
pepesza_ has quit [Excess Flood]
Guest17506 has joined #ipfs
pepesza_ has joined #ipfs
pepesza_ has quit [Excess Flood]
pepesza_ has joined #ipfs
pepesza_ has quit [Excess Flood]
pepesza_ has joined #ipfs
pepesza_ has quit [Excess Flood]
tmg has joined #ipfs
mildred has quit [Remote host closed the connection]
mildred has joined #ipfs
pfraze has quit [Remote host closed the connection]
tmg has quit [Ping timeout: 245 seconds]
m0ns00nfup has joined #ipfs
tmg has joined #ipfs
The_8472 has quit [Ping timeout: 264 seconds]
m0ns00nfup has quit [Quit: undefined]
tmg has quit [Quit: leaving]
tmg has joined #ipfs
Guest17506 has quit [Ping timeout: 245 seconds]
zz_r04r is now known as r04r
pepesza_ has joined #ipfs
pepesza_ has quit [Client Quit]
pepesza has joined #ipfs
slothbag has joined #ipfs
rendar has joined #ipfs
jhulten_ has quit [Ping timeout: 272 seconds]
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #233: Update lodash to version 4.2.0
<daviddias>
noffle: what is the case where you have your pubkey and it is expired? If we want to make it generic, and we a resolving any pubkey and the record that yields is expired, that means that by IPRS spec for that type of record, that data is not valid anymore. We might error and have a force option, but by default we shouldn't resolve data to the user that is
<daviddias>
outdated by the validity metric
<Kubuxu>
daviddias: problem isn't with validity metric but with deciding to look for newer records in the network
Poefke has joined #ipfs
<ipfsbot>
[webui] dignifiedquire pushed 1 new commit to master: https://git.io/vgTkI
blepharosyndesmi has quit [Remote host closed the connection]
jaboja has joined #ipfs
undergrounder has joined #ipfs
zz_r04r is now known as r04r
bauruine has joined #ipfs
jhulten_ has joined #ipfs
jhulten_ has quit [Ping timeout: 240 seconds]
ulrichard has joined #ipfs
hellertime has joined #ipfs
ale___ has joined #ipfs
ale___ has quit [Quit: Page closed]
bauruine has quit [Read error: Connection reset by peer]
elima has quit [Ping timeout: 265 seconds]
zorglub27 has joined #ipfs
IlanGodik_ has joined #ipfs
NeoTeo_ has joined #ipfs
shadoxx has joined #ipfs
lgierth_ has joined #ipfs
IlanGodik has quit [Ping timeout: 240 seconds]
IlanGodik_ is now known as IlanGodik
NeoTeo has quit [Ping timeout: 240 seconds]
ljhms has quit [Ping timeout: 240 seconds]
shadoxx_ has quit [Ping timeout: 240 seconds]
noffle has quit [Ping timeout: 240 seconds]
ffmad_ has quit [Ping timeout: 240 seconds]
lgierth has quit [Ping timeout: 240 seconds]
NeoTeo_ is now known as NeoTeo
pinbot has quit [Ping timeout: 240 seconds]
feross has quit [Ping timeout: 240 seconds]
noffle has joined #ipfs
pinbot has joined #ipfs
feross has joined #ipfs
ljhms has joined #ipfs
__konrad_ has joined #ipfs
ffmad_ has joined #ipfs
elima has joined #ipfs
bauruine has joined #ipfs
reit has joined #ipfs
bauruine has quit [Read error: Connection reset by peer]
m0ns00nfup has quit [Quit: undefined]
tlevine has quit [Ping timeout: 245 seconds]
m0ns00nfup has joined #ipfs
m0ns00nfup has quit [Client Quit]
computerfreak has quit [Remote host closed the connection]
undergrounder has quit [Ping timeout: 240 seconds]
Myagui has quit [Ping timeout: 250 seconds]
<daviddias>
Kubuxu: you would query for records until a timeout ocours or you get a very strong degree of confidence that you have the must updated record that is also still valid. If you are offline and you have an invalid record, the nothing would get resolved, if you are offline and have a valid record, but it is not the most recent one, you would resolve it and it
<daviddias>
is true that you wouldn't be seeking the most recent version, but this isn't a new problem, it is similar when you have to wait for dns to propagate
<daviddias>
But in ipns case, at least you can still resolve offline the previous and still valid version, which is pretty sweet
m0ns00nfup has joined #ipfs
m0ns00nfup has quit [Client Quit]
strg has joined #ipfs
Myagui has joined #ipfs
slothbag has quit [Quit: Leaving.]
bauruine has joined #ipfs
m0ns00nfup has joined #ipfs
Akaibu has joined #ipfs
cryptix has joined #ipfs
bauruine has quit [Read error: Connection reset by peer]
m0ns00nfup has quit [Remote host closed the connection]
<haad>
whyrusleeping: have you been able to work on keystore or pubsub the past couple of weeks?
sevcsik has joined #ipfs
m0ns00nfup has joined #ipfs
gaboose has joined #ipfs
jhulten_ has joined #ipfs
ecloud has quit [Ping timeout: 240 seconds]
m0ns00nfup has quit [Remote host closed the connection]
jhulten_ has quit [Ping timeout: 272 seconds]
G-Ray has quit [Quit: Konversation terminated!]
m0ns00nfup has joined #ipfs
zorglub27 has quit [Ping timeout: 250 seconds]
cemerick has joined #ipfs
m0ns00nfup has quit [Remote host closed the connection]
m0ns00nfup has joined #ipfs
pfraze has joined #ipfs
Encrypt has quit [Quit: Quitte]
m0ns00nfup has quit [Remote host closed the connection]
jaboja has quit [Ping timeout: 240 seconds]
disgusting_wall has joined #ipfs
m0ns00nfup has joined #ipfs
m0ns00nfup has quit [Client Quit]
m0ns00nfup has joined #ipfs
JasonWoof has joined #ipfs
JasonWoof has joined #ipfs
arpu has quit [Remote host closed the connection]
ashark has joined #ipfs
The_8472 has joined #ipfs
m0ns00nfup has quit [Remote host closed the connection]
m0ns00nfup has joined #ipfs
mildred has quit [Quit: Leaving.]
mildred has joined #ipfs
ylp has quit [Quit: Leaving.]
simonv3 has joined #ipfs
trhm has joined #ipfs
trhm has left #ipfs ["Leaving"]
Guest17506 has joined #ipfs
Guest17506 has quit [Ping timeout: 260 seconds]
m0ns00nfup has quit [Remote host closed the connection]
<Kubuxu>
Thanks to [@Kubuxu](//github.com/Kubuxu), dnslink was extended by dedicated subdomain (`_dnslink.`). This allows users to alias (using CNAME) the domain to gateway.ipfs.io while still being able to to set the dnslink to content they wish.
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed feat/mfs-flush-cmd from 8bf9b6d to 0f379c3: https://git.io/vzvUK
<ipfsbot>
go-ipfs/feat/mfs-flush-cmd 161eea7 Jeromy: flushing and shallow list names...
<ipfsbot>
go-ipfs/feat/mfs-flush-cmd 24eb106 Jeromy: use correct context in pubfunc pinning...
<Kubuxu>
richardlitt: ^^
ulrichard has quit [Read error: Connection reset by peer]
<Kubuxu>
richardlitt: once more change: Thanks to [@Kubuxu](//github.com/Kubuxu), dnslink was extended by dedicated subdomain (`_dnslink.`). This allows users to alias (using CNAME) their domain to gateway.ipfs.io while still being able to to set the dnslink to content they wish.
<richardlitt>
thanks Kubuxu. added
<richardlitt>
<3
<richardlitt>
I will send you stickers if you wish, as payment for this mighty gift.
<ipfsbot>
[go-ipfs] RichardLitt created docs/remove-extra-space (+1 new commit): https://git.io/vgk5u
<ipfsbot>
go-ipfs/docs/remove-extra-space 64e5a95 Richard Littauer: Removed extra space...
<ipfsbot>
[go-ipfs] RichardLitt opened pull request #2282: Removed extra space (master...docs/remove-extra-space) https://git.io/vgk52
<ansuz>
lol
<ansuz>
I don't find pgp that hard
<ansuz>
idk
* ansuz
1337
<Kubuxu>
I use netcat over cjdns.
<ansuz>
:D
<ansuz>
same
<daviddias>
richardlitt: <3, I just try to make sure you can spend less time chasing our tails and doing all the awesome things you do for IPFS and the community :)
<ansuz>
nc6 -lp 12345
patcon has joined #ipfs
<ansuz>
nc6 -lp 12345 > somePass.txt
<Kubuxu>
I feel bad because I leak hypeIRC talks over clearnet sometimes (I don't have cjdns on my phone).
<ipfsbot>
[go-ipfs] Kubuxu opened pull request #2283: Make non recursive resolve print the result (master...fix/resolve-norec) https://git.io/vgkhv
m0ns00nfup has quit [Client Quit]
<whyrusleeping>
fuckifiknow
* whyrusleeping
doesnt do http stuff
m0ns00nfup has joined #ipfs
<daviddias>
richardlitt: apparently, that's what the APi expect, but to make sure, you can use wireshark and eavesdrop one of those http requests from the CLI to the API
m0ns00nfup has quit [Remote host closed the connection]
<Kubuxu>
daviddias: ipscend possibly broke on ipfs04
<patagonicus>
I have a dedicated server that's paid for until 2016-03-09 - anything useful I can do for IPFS with that?
<richardlitt>
daviddias: that doesn't work, either.
<daviddias>
Kubuxu: it is possible, js-ipfs-api has changed considerably and even the IPFS API changes a bit on 0.4.0 (and we have a js-ipfs-api branch for it)
m0ns00nfup has joined #ipfs
<daviddias>
Kubuxu: woa that is strange
<daviddias>
it should still add the right folder
<daviddias>
Kubuxu: I want to do an ipscend full revisit once we land 0.4.0
<Kubuxu>
It adds that folder, but I have no idea what it does with it.
<daviddias>
richardlitt: what does wireshark shows you?
<daviddias>
changing config values through the cli or API has been historically no intuitive, I remember having spent some time once just to set some headers
<richardlitt>
nothing
<richardlitt>
Wireshark is showing me nothing.
<daviddias>
when you do an ipfs config bootstrap "bananas" --json?
<daviddias>
s/bootstrap/Bootstrap
<ipfsbot>
[go-ipfs] whyrusleeping deleted docs/add-quotes-to-config at 548ad10: https://git.io/vgIvE
<richardlitt>
Using the filter: "tcp.port == 5001", I get a constant stream
<richardlitt>
Not sure why, I used to be able to only see requests I was making
<ipfsbot>
[go-ipfs] Kubuxu opened pull request #2284: Change smaple API origin settings from wildcard to example.com (master...fix/#2035) https://git.io/vgI5i
ugjka has quit [Ping timeout: 250 seconds]
patcon has quit [Ping timeout: 245 seconds]
<richardlitt>
mmm smaple
<Kubuxu>
richardlitt: shhh, fixing it now
maxlath has joined #ipfs
<Kubuxu>
fixed
zorglub27 has quit [Ping timeout: 260 seconds]
maxlath is now known as zorglub27
kaiza has joined #ipfs
Encrypt has joined #ipfs
kandinsk1 is now known as kandinski
edrex has quit [Read error: Connection reset by peer]
<whyrusleeping>
dignifiedquire: but the less important stuff blocks the important stuff :/
zorglub27 has quit [Ping timeout: 240 seconds]
<dignifiedquire>
that is unfortunate :/
jhulten_ has quit [Ping timeout: 240 seconds]
Kane` has quit [Ping timeout: 250 seconds]
<Kubuxu>
lidel: thanks for your work on FF add-on, it is just awesome
<whyrusleeping>
lgierth_: ping
patcon_ has joined #ipfs
Matoro has quit [Ping timeout: 245 seconds]
patcon has quit [Ping timeout: 240 seconds]
Matoro has joined #ipfs
zorglub27 has joined #ipfs
IlanGodik has quit [Quit: Connection closed for inactivity]
<voxelot>
so i figured out how to get the public key (modulus and exponent) from the format stored in the config file (base64 protobuf of DER bytes)
O47m341 has joined #ipfs
<voxelot>
but i can't get the public key to exactly match after remarshal and encoding.. guess go-ipfs will just have to change the format for the private key and change everyones ID hash :D
redfish has joined #ipfs
redfish has quit [Client Quit]
b0at has quit [Quit: b0at]
Oatmeal has quit [Ping timeout: 248 seconds]
ashark has quit [Ping timeout: 245 seconds]
<Shibe>
is ipfs backwards compatible?
<Shibe>
like can ipfs 4.0 client connect to another ipfs 3.x client?
<voxelot>
Shibe: v0.4.0 can't talk to the older clients
<Shibe>
ok
Matoro has quit [Ping timeout: 245 seconds]
jhulten_ has joined #ipfs
<Codebird>
ipfs daemon - success.
<Codebird>
nohup ipfs daemon & - crash horribly.
<Codebird>
Weird.
elima has quit [Ping timeout: 245 seconds]
<daviddias>
whyrusleeping: ahahah
<Codebird>
Is 0.4.0 done yet?
hellertime has quit [Quit: Leaving.]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
<Codebird>
Never mind. Can't think. Must sleep.
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to feat/debug-events: https://git.io/vgLEO
<ipfsbot>
go-ipfs/feat/debug-events 0db678b Jeromy: log dialing of providers...
* whyrusleeping
wishes he could be productive today
<daviddias>
Kubuxu: whyrusleeping is it my brain, or there is always a white line on blue line connection the 2 vertices on the left
<daviddias>
and also, it is more clear on the black and white versions
<whyrusleeping>
daviddias: i think youre crazy
<daviddias>
is that intended?
<Kubuxu>
daviddias: on vector or raster (png)?
<whyrusleeping>
i see the white line on the black and white one
<Kubuxu>
there is one pixel line on original logo (png0
<Kubuxu>
best seen on 1024x
<daviddias>
Kubuxu: was that intended ? I'm not sure
<Kubuxu>
no it wasn't
<Kubuxu>
only remade by me are vector versions, I will just raster all those bitmaps once we decide what we want precisely
<noffle>
hey jbenet! when we resolve for a key we default to requiring a dht query response from 16 (hardcoded) sources. the local dht can count as one. AFTER we achieve this we validate the record (EOL, PK, etc). sound right so far?
zorglub27 has quit [Quit: zorglub27]
<jbenet>
requirements are right. can start validating the records beforehand. the point is to get 16 _valid_ records. or Exhaust the dht query
kaiza has joined #ipfs
<noffle>
jbenet: *unless* we have no peers -- we count that as an error
<jbenet>
consistency here is hard. for something like naming, it's a BFD if this isn't consistent. for something like a provider record, consistency is not a big deal