[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #23308 [Core Tor/Tor]: prop224: Implement note_connection_attempt_succeeded() in the client code
#23308: prop224: Implement note_connection_attempt_succeeded() in the client code
-----------------------------+------------------------------------
Reporter: dgoulet | Owner: dgoulet
Type: enhancement | Status: needs_review
Priority: High | Milestone: Tor: 0.3.2.x-final
Component: Core Tor/Tor | Version:
Severity: Normal | Resolution:
Keywords: tor-hs, prop224 | Actual Points:
Parent ID: #23300 | Points:
Reviewer: asn | Sponsor: SponsorR-must
-----------------------------+------------------------------------
Changes (by dgoulet):
* status: assigned => needs_review
* reviewer: => asn
Comment:
See branch: `ticket23308_032_01`.
I *think* it's the cleanest approach I could find. The problem is not that
trivial for the edge connection identifier to keep the blinded key it has
used when sending the fetch to the directory. The problem is that this
whole chain is async and we don't pass the edge connection ident through
the set of functions to fetch a descriptor.
Furthermore, the time period is also taken when the fetch is triggered so
we can't really put that value in the edge conn identifier either.
So, middle ground is to try to purge for both time period.
I'm really open to a new solution if one could be cleaner but definitely
passing the edge connection up to the fetch function would be messy and
conditional to the context...
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/23308#comment:1>
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