[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #5646 [Tor Client]: Many functions log before checking for proto violation
#5646: Many functions log before checking for proto violation
------------------------+---------------------------------------------------
Reporter: asn | Owner:
Type: defect | Status: new
Priority: normal | Milestone: Tor: 0.2.3.x-final
Component: Tor Client | Version:
Keywords: | Parent: #5643
Points: | Actualpoints:
------------------------+---------------------------------------------------
Comment(by asn):
Replying to [comment:1 arma]:
> When we are changing this, it would be great to log whatever we used to
log when declaring the protocol violation. Part of the reason those
initial logs are there is to let people reading the logs follow what is
happening (even if it shouldn't be happening!)
OK. This means that in cases like `rend_mid_introduce()`, we also have to
print `p_circ_id` in the protocol violation log, which is not too
different from the current situation. If we consider printing `p_circ_id`
totally innocent, then maybe we should leave such code as is, instead of
tweaking it to add the `p_circ_id` in the protocol violation log.
If the above paragraph sounds like a good idea, then the only offender of
this ticket that I can find is the:
{{{
reason = (uint8_t)cell->payload[0]
}}}
in `command_process_destroy_cell()`:
{{{
circ = circuit_get_by_circid_orconn(cell->circ_id, conn);
reason = (uint8_t)cell->payload[0];
if (!circ) {
log_info(LD_OR,"unknown circuit %d on connection from %s:%d.
Dropping.",
cell->circ_id, conn->_base.address, conn->_base.port);
return;
}
}}}
--
Ticket URL: <https://trac.torproject.org:443/projects/tor/ticket/5646#comment:2>
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