ndt-users list created

[Jeremian <>]

>> This is a bug in the Web100 library. Please look at
>> http://trinity.web100.org/pipermail/discussion/2006-07/msg00008.html
>>
>> Unfortunately they haven't released the fixed version yet. However,
>> the needed patch is very small and easy to apply.
>
>Yes, thanks for the pointer! When trying to apply the patch, I also
>noticed that the NDT server was linked against an alpha version (1.4)
>of the Web100 userspace library... which may explain other issues.
>
>I moved to 1.5, applied the patch, recompiled the NDT server, and now
>IPv4 mode works just fine with my Solaris web100clt:
>

[...]

>
>But (native) IPv6 still gives the same warning about MSS:
>

[...]

>
>It's normal that the patch doesn't help here, since it is only
>concerned with mapped addresses, and in my test, "real" IPv6
>addresses are used.

I forgot that the patch in the mentioned mail wasn't complete. Here
comes the missing part:

 /-----------------------------------------------------------------\
@@ -703,7 +703,7 @@

     cp = agent->info.local.connection_head;
     while (cp) {
-        if (memcmp(&cp->spec_v6, spec_v6, sizeof (spec_v6)) != 0)
+        if (memcmp(&cp->spec_v6, spec_v6, sizeof (spec_v6)) == 0)
             break;
         cp = cp->info.local.next;
     }
 \-----------------------------------------------------------------/

Sorry for the trouble.

>>> And I still have an issue using the client over IPv4.
>>>
>>> : 
>>> leinen@diotima[src];
>>>  ./web100clt -4 -n lsmp2
>>> Testing network path for configuration and performance problems
>>> -- Using IPv4 address
>>> Checking for Middleboxes . . . . . . . . . . . . . . . . . .
>>> Protocol error!
>>> : 
>>> 1leinen@diotima[src];
>>>
>>> I'll dig further into this.  I already checked that I can telnet
>>> to the IPv4 address of "lsmp2".
>>
>> Can you provide the server's output with the higher debug level
>> (7)?
>
>I could reproduce the "Protocol error!" situation, but this time with
>IPv6.  It seems to happen a bit arbitrarily.  I'm attaching two
>server logs from `web100srv -a -ddddddd'.  The first is from a
>"Protocol error!" run, the second from a successful one (but
>mis-guessing the MSS as 16376).

For some reasons the timer (15 seconds) has expired in the Middlebox
test. Can you provide the client's output with the same high debug
level? This will be helpful in investigating the problem.

Jakub.