[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #5083 [Obfsproxy]: Implement heartbeat message in obfsproxy
#5083: Implement heartbeat message in obfsproxy
-------------------------+--------------------------------------------------
Reporter: karsten | Owner: asn
Type: enhancement | Status: needs_review
Priority: normal | Milestone:
Component: Obfsproxy | Version:
Keywords: | Parent:
Points: | Actualpoints:
-------------------------+--------------------------------------------------
Changes (by karsten):
* status: needs_revision => needs_review
Comment:
Replying to [comment:7 karsten]:
> Replying to [comment:6 asn]:
> > The check
> > {{{
> > p = strrchr(addrport, ':');
> > if (p < 0)
> > return;
> > }}}
> > in `status_note_connection()` seems wrong. `strrchr()` returns a
pointer, so `if (!p)` is probably better than `if (p < 0)`. I wonder if we
should use `strchr()` like `util.c:resolve_address_port()` does; but it
probably doesn't really matter.
>
> Ah, there's the code I was looking for. I'm going to steal it from
there.
Fixed.
Note that strchr() does the wrong thing here, because it finds the first
':' in the string, not the last. That doesn't work for IPv6 addresses.
That's a bug in resolve_address_port() in util.c, too!
> > b) Should we use a hash table or something to query whether a
connection has been seen before? It seems that `smartlist_string_isin()`
does a linear search over a smartlist, and some obfsproxy bridge operators
are currently seeing 1500~ or so unique IPs. I'm not sure if that would
cause lots of load, but a hash table (I think that's a `strmap_t` in
`container.c`) might be better.
>
> Oh. You're right. strmap_t it is.
Fixed.
Please see my updated branch heartbeat_uniqueips.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/5083#comment:9>
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