[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #27471 [Core Tor/Tor]: HS intermittently fails: Non-fatal assertion failed in send_introduce1
#27471: HS intermittently fails: Non-fatal assertion failed in send_introduce1
---------------------------------+------------------------------------
Reporter: tgragnato | Owner: dgoulet
Type: defect | Status: accepted
Priority: Very High | Milestone: Tor: 0.3.5.x-final
Component: Core Tor/Tor | Version: Tor: 0.3.4.7-rc
Severity: Minor | Resolution:
Keywords: tor-hs, regression? | Actual Points:
Parent ID: | Points:
Reviewer: | Sponsor:
---------------------------------+------------------------------------
Comment (by asn):
Replying to [comment:4 dgoulet]:
> I'm still unsure why this happened but it is clear that the RP/IP
circuit were established. Once the RP ACKed the request, the
authentication key in the IP `hs_ident` wasn't the one in the descriptor.
>
> So between the initial requests for that .onion and the ACK from the RP,
the descriptor simply changed and the tor client fetched/stored it.
>
> That makes me think that when we store a *new* HS descriptor for A.onion
that is we already had one in our cache and we are getting a new one, we
should probably close every circuits related to the old one? Probably
something like going over the IP in the old one, if they are different
from the new one, close any related circuits.
>
> That way, we don't end up with leftover circuit for the *same* A.onion
but with outdated data.
I'm good with the above "close obsolete circuits" fix as long as it's not
complicated (ideally just a few simple lines of code). If it ends up being
complicated code, we should consider other approaches, or maybe try to
reproduce it.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/27471#comment:6>
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