client bug in and a new bad exit: exoassist

grarpamp grarpamp at
Fri Feb 5 09:45:07 UTC 2010

Since I doubt the suggested tests were performed, I did the work
instead. I reached the 'up' website, and timed out on the 'down'
one... both as expected. And I diffed the clearnet and tornet
versions of the 'up' one and they matched, sans 'alteration'.

If one can prove exo is the site with the cache, and that it's
also configured improperly, this case might have more merit.

Can people tell me which of all the websites they surfed today had
a cache somewhere between their browser and the site httpd? Did one
block all those internet paths on a whim against caches? Is TorProject
going to test all exits to examine for potential caches in the path
and flag them all for similar reasons? And what exactly does one
propose to do about all the caches and other proxies it can't detect
because they happen to be more transparent?

I highly think not, on the former three accounts. And 'oops, policy
failure' on the latter.

Some cache in the path in question is ejecting a harmless and indeed
rather useful error message. One that even caught a bug in Tor.

Users are free to block exo in their config, or not. In the grand
scheme of things, I'd suggest it's a waste of time.

> security breaches

So now caches are a security breach too? That's news to me, lol :)

It's up to torproject now. Happy Tor'ing as always :)

And since I'm sure someone will bring up headers somewhere in all
this mootness, here's a few of those too :)

GET / HTTP/1.0

HTTP/1.0 200 OK
Content-Type: text/html
Accept-Ranges: bytes
ETag: "xxxxxxxxxx"
Last-Modified: Tue, 02 Feb 2010 13:45:19 GMT
Content-Length: 20987
Date: xxx, xx Feb 2010 xx:xx:xx GMT
Server: httpd/1.4.x LaHonda
X-Cache: MISS from
X-Cache-Lookup: MISS from
Age: xxxx
X-Cache: HIT from
X-Cache-Lookup: HIT from
Via: 1.0 (squid/2.7.STABLE3), 1.0 (squid/2.7.STABLE3)
Connection: close
router exoassist > [not found]
To unsubscribe, send an e-mail to majordomo at with
unsubscribe or-talk    in the body.

More information about the tor-talk mailing list