<whyrusleeping>
yeah, i think the iptb tests are randomly failing on travis due to port conflicts
<whyrusleeping>
i could be wrong
Daboloskov has quit [Client Quit]
tilgovi has quit [Ping timeout: 245 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
tilgovi has joined #ipfs
philholden has joined #ipfs
pfraze has quit [Remote host closed the connection]
lgierth has joined #ipfs
pfraze has joined #ipfs
gatesvp has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
inconshreveable has quit [Remote host closed the connection]
hellertime has quit [Quit: Leaving.]
Daboloskov has joined #ipfs
Daboloskov has quit [Client Quit]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
jabroney has joined #ipfs
besenwesen has joined #ipfs
j-pb has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
gatesvp has quit [Ping timeout: 246 seconds]
j-pb has joined #ipfs
EricJ2190 has joined #ipfs
j-pb has quit [Remote host closed the connection]
j-pb has joined #ipfs
AtnNn has quit [Remote host closed the connection]
<whyrusleeping>
can someone look over my network simulator code?
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to feat/netsim: http://git.io/vJlLT
<ipfsbot>
go-ipfs/feat/netsim 612ca19 Jeromy: a few tweaks to the netsim code
inconshreveable has joined #ipfs
therealplato has joined #ipfs
grncdr has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
grncdr1 has joined #ipfs
grncdr has quit [Ping timeout: 256 seconds]
j-pb has quit [Remote host closed the connection]
lgierth_ has joined #ipfs
Daboloskov has joined #ipfs
Daboloskov has quit [Client Quit]
lgierth has quit [Ping timeout: 276 seconds]
pfraze has quit [Remote host closed the connection]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
j-pb has joined #ipfs
anshukla has joined #ipfs
j-pb has quit [Remote host closed the connection]
j-pb has joined #ipfs
mildred has quit [Quit: Leaving.]
j-pb has quit [Remote host closed the connection]
pfraze has joined #ipfs
notduncansmith has joined #ipfs
<wking>
whyrusleeping: feat/netsim looks reasonable to me, although Go's networking is new to me. I like the transport goroutine. What sort of feedback are you looking for?
notduncansmith has quit [Read error: Connection reset by peer]
j-pb has joined #ipfs
RzR has quit [Excess Flood]
RzR has joined #ipfs
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
grncdr1 has quit [Quit: Leaving.]
headbite has quit [Quit: Leaving.]
therealplato1 has joined #ipfs
therealplato has quit [Ping timeout: 276 seconds]
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping>
wking: i'm looking to make sure that my implementation mimics a real network pretty well
<luzifer>
whyrusleeping: can you name me a quite popular notification service for $mobile used in the US?
<luzifer>
hmm pushbullet is not nice... :( I'd need to create a google account for gobuilder and sign up with that one... no chance to register an app and send on its behalf...
notduncansmith has joined #ipfs
<whyrusleeping>
thats obnoxious...
notduncansmith has quit [Read error: Connection reset by peer]
<luzifer>
just sent them a message asking for advice how to work around this
Daboloskov has joined #ipfs
Daboloskov has quit [Client Quit]
j-pb has quit []
<luzifer>
but first: sleep. gn8 everyone!
<whyrusleeping>
luzifer: gnite!
notduncansmith has joined #ipfs
notduncansmith has quit [Read error: Connection reset by peer]
<whyrusleeping>
Tv`: how does tcp handle things if you write 1 byte 1000 times?
<Tv`>
whyrusleeping: every write(2) syscall (or moral equivalent) puts the data given to it in the send queue (if there's room)
<Tv`>
writing a byte at a time just means you do 1000 syscalls instead of 1
<whyrusleeping>
and how often is the send queue written out?
<Tv`>
it's not a stead tick, it's controlled by the congestion window
<whyrusleeping>
:/
<whyrusleeping>
i'm trying to figure out how to best emulate it
<Tv`>
and whether things get sent or not depends on nagle algorithm etc
<whyrusleeping>
because we have TONS of four byte writes
<Tv`>
unf
pfraze has quit [Remote host closed the connection]
<Tv`>
that rarely changes wire behavior, except for 1 case: it may send a smaller initial segment
<Tv`>
get 4 or 8 or whatever bytes written, rush to send them off VS get 1000 bytes at a time, send as much as congestion window allows
<Tv`>
apart from that, it's just syscall overhead
<Tv`>
-> burning more cpu -> as long as that's not the bottleneck, it's just waste, not really slowdown
<whyrusleeping>
okay... so if i put all writes into an 'outgoing' buffer, and then write it out every N milliseconds, that should be closer to accurate?
anshukla has quit [Remote host closed the connection]
<Tv`>
not really ;)
<whyrusleeping>
fffffffff
<Tv`>
if matching TCP behavior was doable naively, it wouldn't as tricky of a beast