[tor-bugs] #16276 [Onionoo]: bug in onionoo's family set detection
Tor Bug Tracker & Wiki
blackhole at torproject.org
Tue Jul 7 16:53:00 UTC 2015
#16276: bug in onionoo's family set detection
-----------------------------+--------------------------
Reporter: cypherpunks | Owner:
Type: enhancement | Status: needs_review
Priority: normal | Milestone:
Component: Onionoo | Version:
Resolution: | Keywords:
Actual Points: | Parent ID:
Points: |
-----------------------------+--------------------------
Comment (by karsten):
Replying to [comment:39 leeroy]:
> Karsten, are you sure we're talking about the same ff member data? Why
would GSON be setting it? That's the task of metrics-lib not GSON. GSON is
used to construct the JSON response isn't it?
SummaryDocument is an internal document used only by Onionoo and is
serialized using Gson. metrics-lib's task is to parse descriptors
produced by Tor relays and bridges and provided by CollecTor as input to
Onionoo.
But I wasn't entirely right above. I overlooked that the new code would
write SummaryDocuments without the ff attribute, so even if Gson is
capable of deserializing objects with that attribute, if the attribute's
value is null, it's null. Pushed a fix to my branch.
> The cardinality of effective_family for the relay should be 4.
Hmm?
> > Does onionoo's view on effective family match a tor clients view?
>
> If it's working, Onionoo's view is ideal. The tor client uses family as
a hint for path selection. Meaning, assuming you have exactly these latest
descriptors, and your client can connect to all the relay, the path
selection builds bi-directional relationships for you during circuit build
time. So no, in general, it doesn't.
I'd still say that Onionoo's view is a reasonable approximation to a
client's view.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/16276#comment:40>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
More information about the tor-bugs
mailing list