logs archiveIRC Archive / Oftc / #tor / 2010 / February / 8 / 1
flameon
reconnecting gives count 1, and then count 3
no other warnings
nsa
or: atagar committed revision 21580 (/arm/trunk): Had enough of a siesta - getting back into development beginning with a rewrite if the starter.
or: added: made authentication a little smarter, using PROTOCOLINFO to autodetect authentication type and cookie location
or: change: made 'blind mode' (disables connection queries) a startup option rather than flag in source (request by Sebastian)
or: change: all log events (including arm) are now set via character flags, with TorCtl events log as their own toggleable type
or: change: starting log label with runlevel events, condensing if logging a range
or: change: simplifying command line parsing via getopt
atagar
Sebastian: got cut off but one of the entries was "fix: blind mode now prevents all netstats (including connection counts and halting resolver thread), improving performance" in case ya want another argument for people to use the option ;)
Sebastian
ok
I don't care about performance ;)
waltman
My tor bridge has 27 sockets open to softbank218128246013.bbtec.net, all on different ports. Is this normal?
Oh, and it reached its soft bandwidth limit 13 hours ago and is supposedly "hibernating".
picsoung
Hey :)
how are you ? :)
i got a question about a "
502 Server dropped connection"
i dont understand how to fix it...
try to access to wiki.developers.facebook.com
nsa
or: Sebastian Hahn <sebastian@torproject.org>: 2010-02-07 05:30:55 [tor/maint-0.2.1]: lookup_last_hid_serv_request() could overflow and leak memory
or: Sebastian Hahn <sebastian@torproject.org>: 2010-02-07 05:30:55 [tor/master]: lookup_last_hid_serv_request() could overflow and leak memory
or: Nick Mathewson <nickm@torproject.org>: 2010-02-08 03:34:08 [tor/master]: Merge commit 'origin/maint-0.2.1'
or: sebastian committed revision 21581 (/projects/todo): bug1242 fix was merged, woo
tryagain
i seem to remember Tor was some barbarian in fantasy novels
picsoung
:p
it's a nordic god too, right ?
         

tryagain
now I remember Tor was the publisher of various fantasy/science fiction
waltman
yes, they're a big publisher of genre fiction
I believe the nordic god is Thor, not Tor.
picsoung
right
waltman
it's also a rocky outcrop and a frequent crossword puzzle answer
jr__
this is rather nottorish
ra
hi
has anyone had success running cisco vpn through tor?
even if try to active tcptunneling the client connects to udp port 29747 (which fails of course)
utid8i5
how do I enable logging destination address?
Runa
utid8i5: sorry?
utid8i5
Runa: address where I'm trying connect to.
this IRC for example.
Runa
I know what a destination address is, but I don't quite understand what you want to do
utid8i5
Runa: I'm connected to IRC server right now. But in notice.log last string is a '[warn] Please upgrade!'. I would like to see there address of IRC server where I'm chating right now through tor.
Runa
utid8i5: but tor is designed not to log any information about its users
utid8i5
Runa: then it is hard to control do I really connected through tor, or it is dirrect connection.
murb
utid8i5: if you want to find the irc server you're using try /whois utid8i5 ?
utid8i5
murb: yes, but how about web server?
Runa
utid8i5: check.torproject.org will tell you whether or not you are using tor
utid8i5
I remember that old tor version be able to write somethign like [scrubed]:80...
dun
hi
Runa
hi
nsa
or: phobos committed revision 21582 (/website/trunk/include): change the author to tor, since roger doesn't write everything anymore.
or: phobos committed revision 21583 (/website/trunk/en): update sponsored years. this is what i do on a 4h flight delay.
or: phobos committed revision 21584 (/website/trunk/en): move tor-assistants to the top of the list since most people ignore the
or: descriptions anyway.
         

Runa
phobos: \o/
nsa
or: phobos committed revision 21585 (/website/trunk/en): add a link to archive.torproject.org. suggested by barkerjr.
or: phobos committed revision 21586 (/website/trunk/en): s/webmaster/contact us/ to make it more clear
or: phobos committed revision 21587 (/website/trunk/en): point to the rpm repos instead of the individual files.
or: phobos committed revision 21588 (/website/trunk/docs/en): add instructions for linux distributions with rpm packaging.
phobos
now it's a 10h flight delay
(Action) converts the website to php, javascript, and flash
Runa
haha :)
lots of snow?
phobos
no, dead computers
and they re-initialized the entire plane and it still failed
Runa
:o
phobos
now let's see if the site builds
BarkerJr
I was watching this tv show called Human Target... in there, they replaced the plane's computer with a laptop :)
dr|z3d
I was about to call "off-topic!", then I noticed phobos inititiated said dicussion.
(Action) chuckles.
(Action) retrofits an "s"
Tas
they did run Tor on that laptop ;-)
dr|z3d
Ah, then we've come full circle. :) As you were!
nsa
or: phobos committed revision 21589 (/website/trunk/en): link directly to the package dir so people aren't confused.
BarkerJr
hehe
phobos
my offtopicness was bout the 5 commits i did
nsa
or: phobos committed revision 21590 (/website/trunk/en): it appears incognito livecd is process of being replaced by amnesia.
or: link to the latter.
anonym
phobos: I can confirm your suspicion re: incognito being "replaced" with amnesia
phobos
i haven't tried it yet
but great that a livecd lives
i use them when i travel and have to use strange hardware
anonym
it still has some rough edges, but it works pretty well
phobos
it's based on debian, which is 100% improvement over gentoo ;)
anonym
well, to each his own, but it'll likely results on more frequent updates
eh, "result in"
phobos
just harassing you, i don't care which distro it is
anonym
sure, no offense taken
hoj44
hello
Sebastian
hello
nsa
or: sebastian committed revision 21591 (/projects/todo): branch to refactor mlockall
waltman
I'm running a bridge, and when I run netstat I see that over half the connections (41/75) are to a single host. They're all on different large ports between 49000 and 65000. Is this normal, or is something wrong?
since they're all high ports it must be a relay, right?
atagar_work
waltman: assuming they're large ports on your side but most likely a single port on the other side then yes (that indicates a relay) - though it does seem odd that so many are to a single host....
waltman
it's softbank218128246013.bbtec.net
atagar_work
not familiar with it - if you're running arm it'll attempt to correlate against consenus data to tell you more
waltman
what's arm?
atagar_work
www.atagar.com/arm
waltman
Thanks. I'll take a look at that, but I won't have time until tonight.
atagar_work
good luck, let me know if you have any issues with it
waltman
is there a debian package for it?
atagar_work
nope, but instillation is dead simple
waltman
ok
Sebastian
could simply be many people connecting to you from behind a nat
atagar_work
Sebastian: If they were inbound connections they'd all be to his ORPort, yes? Perhapse I need more coffee...
Sebastian
no
erm
yes
but did he say inbound?
atagar_work
he said netstat connection listing - both inbound and outbound (I'm just saying that since these are diverse connections they're probably outbound)
and if they were to a single host on diverse ports that *would* be weird
most likely he's confusing non-tor traffic
Sebastian
no idea
alternatively, the bridge allows one-hop exits
atagar_work
Hu? That's an option?
I thought numerous proposals were trying to nip that in the bud
Sebastian
See the AllowSingleHopExits config optoin
defaults to false, of course
waltman
They're all on ports like 55590
Sebastian
waltman: did you set ExitPolicy reject *:* in your config?
If not, do that
waltman
I yes, that's set
s/I//
netstat says those connections all belong to tor
atagar_work
if they're all on one or two ports to the single host that make sense, if not then I'd be curious what arm has to say about them
waltman
I what am I looking for in arm?
atagar_work
it'll give a connection listing with details from the consensus - if it can't find any consensus data I'd start getting suspicious, otherwise things should then make sense
for instance it'll say that host's OR ports (or possibly mulitple ORPorts)
waltman
how do I see a connection listing?
atagar_work
there's a few (like moria) that have multiple but I've never seen more than a couple
what do you mean?
oops, sorry (thought you said 'how long') - it's a page in the monitor that's periodically refreshed
see screenshots
waltman
the second screen shot looks like what i want. What do I press to get it?
atagar_work
when started you press left and right to switch pages
waltman
got it
You're right. They're all inbound connections.
from somewhere in japan
so it probably is a nat.
so someone's client going crazy
Sebastian
right. that guy's client is probably either going crazy or trying something malicious.
ilter
Hello. Could you explain how Tor decide which OR is taken fast flag?
Sebastian
http://gitweb.torproject.org//tor.git?a=blob_plain;f=doc/spec/dir-spec.txt;hb=HEAD section 3.3
ilter
Thank you Sebastian. I'm reading it.
Path-spec says only for 'fast' circuits Tor picks a given router as an exit with probability proportional to its bw. Could you explain why Tor do it only for 'fast' circuits?
Sebastian
non-fast nodes are so slow that clients will be pretty sad if they try to push traffic through them
waltman
atagar_work: what does the time mean on the connections list in page 2?
how long it's been active? how long it's been idle?
15
meh
atagar_work
waltman: observed uptime of the connection (time that the connection's been alive while arm was running)
ilter
Sebastian: Yes i see but why Tor doesn't pick an exit router with probability proportional to its bw for non-fast circuit? Path-spec says Tor picks an exit router randomly for circuits that don't need to be 'fast'.
waltman
atagar_work: thanks
Sebastian
ilter: dunno. You could ask or-dev
ilter
Ok Sebastian thank you for your concern.
nsa
or: n8fr8 committed revision 21592 (/projects/android/trunk): test
n8fr8
Sebastian: looks like its working. thx.
Sebastian
sure it's working :)
nsa
or: n8fr8 committed revision 21593 (/projects/android/trunk/Orbot): migrating Orbot codebase from old repo and updating to 0.0.3a
or: n8fr8 committed revision 21594 (/projects/android/trunk/Orbot/res/drawable): adding new image resource to Orbot
or: n8fr8 committed revision 21595 (/projects/android/trunk/Orbot/res): adding new/missing string resources
« prev next »