Dating xkubu it

I'm not (yet) convinced that it's a problem with DNSB.

If there's not a problem, there's nothing for Steve to fix.

a rather foolproof procedure to me :/ From the packet capture, I see a dns request first for: s4ojgvvre5y5hn3h3org When DNSB is working properly, this A record request is answered by 10.x.x.x I am presuming when it is not working well, for whatever reasons, 10.x.x.x does not respond at all...

If that is an initial "all or nothing" test, (is that what Steve is saying) then I guess it doesn't pass mustard *at that point in time*.

(It is confusing) The query that would go to any locally-configured resolvers would not be for that whacky & deliberately bogus sub-domain name ...

The latest build also fails for me under what can only be described as some sort of bizarre corporate network setup (mine!

) Anyway, as posted in that other thread, this is likely of academic interest for SG since it is very likely something off the wall here... [for the unabridged version, see Kerry Liles's post above] FWIW, Kerry, I *DID* know that these recent changes would not be solving your long-standing corporate proxying problem.

Essentially, THAT trouble you are having is "by design"...

DNSB displays the screen "Determining Internet Connectivity", but it really means: "Determining connectivity that's useful for performing the benchmarking that DNSB was created to perform." So...

I regard that what DNSB is doing is actually CORRECT behavior since, as we've discussed before, behind that restrictive proxy, DNSB can't really do anything other than test the performance of the one or two local resolvers that it's allowed to access.

Leave a Reply