We've got millisecond timers now, we might as well use them.
This change won't actually make circuits get expiered with microsecond
precision, since we only call the expiry functions once per second.
Still, it should avoid the situation where we have a circuit get
expired too early because of rounding.
A couple of the expiry functions now call tor_gettimeofday: this
should be cheap since we're only doing it once per second. If it gets
to be called more often, though, we should onsider having the current
time be an argument again.
When calling circuit_build_times_shuffle_and_store_array, we were
passing a uint32_t as an int. arma is pretty sure that this can't
actually cause a bug, because of checks elsewhere in the code, but
it's best not to pass a uint32_t as an int anyway.
Found by doorss; fix on 0.2.2.4-alpha.
Left circuit_build_times_get_bw_scale() uncommented because it is in the wrong
place due to an improper bug2317 fix. It needs to be moved and renamed, as it
is not a cbt parameter.
To quote arma: "So instead of stopping your CBT from screaming, you're just
going to throw it in the closet and hope you can't hear it?"
Yep. The log message can happen because at 95% point on the curve, we can be
way beyond the max timeout we've seen, if the curve has few points and is
shallow.
Also applied Nick's rule of thumb for rewriting some other notice log messages
to read like how you would explain them to a raving lunatic on #tor who was
shouting at you demanding what they meant. Hopefully the changes live up to
that standard.
We need to make sure that the worst thing that a weird consensus param
can do to us is to break our Tor (and only if the other Tors are
reliably broken in the same way) so that the majority of directory
authorities can't pull any attacks that are worse than the DoS that
they can trigger by simply shutting down.
One of these worse things was the cbtnummodes parameter, which could
lead to heap corruption on some systems if the value was sufficiently
large.
This commit fixes this particular issue and also introduces sanity
checking for all consensus parameters.
We really should ignore any timeouts that have *no* network activity for their
entire measured lifetime, now that we have the 95th percentile measurement
changes. Usually this is up to a minute, even on fast connections.
If we really want all this complexity for these stages here, we need to handle
it better for people with large timeouts. It should probably go away, though.
Rechecking the timeout condition was foolish, because it is checked on the
same codepath. It was also wrong, because we didn't round.
Also, the liveness check itself should be <, and not <=, because we only have
1 second resolution.
Specifically, a circ attempt that we'd launched while the network was
down could timeout after we've marked our entrynodes up, marking them
back down again. The fix is to annotate as bad the OR conns that were
around before we did the retry, so if a circuit that's attached to them
times out we don't do anything about it.
We frequently add cells to stream-blocked queues for valid reasons
that don't mean we need to block streams. The most obvious reason
is if the cell arrives over a circuit rather than from an edge: we
don't block circuits, no matter how full queues get. The next most
obvious reason is that we allow CONNECTED cells from a newly created
stream to get delivered just fine.
This patch changes the behavior so that we only iterate over the
streams on a circuit when the cell in question came from a stream,
and we only block the stream that generated the cell, so that other
streams can still get their CONNECTEDs in.
