[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #15618 [Core Tor/Tor]: Tried to establish rendezvous on non-OR circuit with purpose Acting as rendevous (pending)
#15618: Tried to establish rendezvous on non-OR circuit with purpose Acting as
rendevous (pending)
-------------------------------------------------+-------------------------
Reporter: asn | Owner: dgoulet
Type: defect | Status:
| needs_information
Priority: Medium | Milestone: Tor:
| unspecified
Component: Core Tor/Tor | Version:
Severity: Normal | Resolution:
Keywords: tor-hs needs-insight needs- | Actual Points:
diagnosis |
Parent ID: | Points: 1
Reviewer: | Sponsor:
-------------------------------------------------+-------------------------
Comment (by dgoulet):
I've made this reproducible but with a change to the tor code... I wanted
to test my theory that basically at least _two_ `ESTABLISH_RENDEZVOUS`
cells are sent to the RP on the same circuit. I've modified Tor to pin a
RP I control and then made it call twice
`rend_client_rendcirc_has_opened()` in `circuit_has_opened()`.
And here you go, you can see this log being triggered on my relay (the
RP):
{{{
Jul 02 18:31:47.000 [warn] Tried to establish rendezvous on non-OR circuit
with purpose Acting as rendevous (pending)
}}}
This means somehow that either the tor client can call twice
`circuit_has_opened()` and which means it can send two cells on the same
circuit (#21084) or then some other client implementation out there is
doing the wrong thing triggering warning everywhere.
We should probably downgrade that log_warn to a PROTOCOL warning because
anyone out there can send two cells on the same circuit just "for fun"...
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/15618#comment:27>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs