questions about bug in 0.2.0.20-rc
Roger Dingledine
arma at mit.edu
Sun Jun 15 02:26:56 UTC 2008
On Sat, Jun 14, 2008 at 08:57:31PM -0500, Scott Bennett wrote:
> I just came home and found tor had crashed. I've been running 0.2.0.20-rc
> lately, which is the first version with the clock jump bug, but it doesn't do
> it very often. However, after this crash, I'm back to 0.2.0.19-alpha for the
> moment. 0.2.0.20-rc crashed with the following message in the info-level log:
>
> Jun 14 19:23:08.176 [err] Bug: connection.c:1580: connection_buckets_decrement: Assertion num_written < INT_MAX failed; aborting.
>
> My questions are:
> 1) What does this bug entail? (I.e., what does the message mean?)
> 2) Has it been fixed in a later version already?
The 0.2.0.21-rc ChangeLog includes:
- Downgrade assert in connection_buckets_decrement() to a log message.
This may help us solve bug 614, and in any case will make its
symptoms less severe. Bugfix on 0.2.0.20-rc. Reported by fredzupy.
Then 0.2.0.22-rc says:
- When counting the number of bytes written on a TLS connection,
look at the BIO actually used for writing to the network, not
at the BIO used (sometimes) to buffer data for the network.
Looking at different BIOs could result in write counts on the
order of ULONG_MAX. Fixes bug 614.
--Roger
More information about the tor-talk
mailing list