[tor-bugs] #10362 [Pluggable transport]: Deploy FTE as a pluggable transport in PTTBBs
Tor Bug Tracker & Wiki
blackhole at torproject.org
Wed Dec 18 01:30:03 UTC 2013
#10362: Deploy FTE as a pluggable transport in PTTBBs
-------------------------------------+-----------------
Reporter: asn | Owner:
Type: task | Status: new
Priority: normal | Milestone:
Component: Pluggable transport | Version:
Resolution: | Keywords:
Actual Points: | Parent ID:
Points: |
-------------------------------------+-----------------
Comment (by asn):
Replying to [comment:7 kpdyer]:
> Replying to [comment:5 dcf]:
> > Replying to [comment:4 kpdyer]:
> > > The changes above introduce the following question: If the inputs to
(un)rank are out of range, then that indicates something has gone
seriously wrong in the system. That is, we shouldn't ever invoke unrank on
an integer that we don't know how to unrank. What's the "proper" PT way to
handle this fatal error condition?
> >
> > I don't think you can do anything except close the connection and
write to a log file somewhere. It's the same as if a MAC failed to verify,
or a WebSocket message were too long.
>
> Is there a standard paradigm for logging from PTs?
>
Not really. But there should be one.
Obfsproxy, when used in managed mode, only allows logging to a file. I
think this is the easiest and "cleanest" solution for now. (This happens
because its stdout channel is used for the managed-proxy configuration
protocol, and its stderr is unused (#9957).). For example, you do:
{{{ClientTransportPlugin obfs2 exec /usr/local/bin/obfsproxy --managed
--log-file=/home/user/logq}}}
Apart from #9957 (which will certainly be helpful), we might be able to
design a better logging solution for PTs by using the Extended ORPort or
the TransportControlPort (unimplemented; see 196). This has not been
particularly needed so far.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/10362#comment:9>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
More information about the tor-bugs
mailing list