[bind10-dev] #1534, IPV6_USE_MIN_MTU and similar

Michal 'vorner' Vaner michal.vaner at nic.cz
Tue Feb 21 11:59:57 UTC 2012


Hello

On Mon, Feb 20, 2012 at 11:17:09AM +0100, Shane Kerr wrote:
> On Monday, 2012-02-20 10:34:20 +0100, 
> Shane Kerr <shane at isc.org> wrote:
> > So it looks like MTU is set per-destination, which means that a
> > connected socket is required. :(
> 
> Michal points out on jabber that one can happily *set* the MTU, just
> *reading* the value is not supported.
> 
> As we discussed on jabber, we can try to connect() during the tests,
> which is where the pain lies.

The current error is following:
[ RUN      ] get_sock.udp6_create
sockcreator_tests.cc:140: Failure
Value of: options
  Actual: 16436
Expected: 1280
[  FAILED  ] get_sock.udp6_create (0 ms)

This means that the setting doesn't work either, it just doesn't return any
error (which is probably bug of somebody else. However, being it bug of somebody
else doesn't solve our problem ‒ what to do in case of Linux. I'd have a
proposal, consisting of three timelines:
• Remove the setting of MTU from this ticket and be happy it works at last
  somewhere.
• Write code that connects the socket before sending, sets MTU, sends and then
  uses the socket to connect to a different place. This should be possible, but
  complicated (because we need to use a different socket that the one listening,
  we need to bind it to the same address (hmm, is it possible? probably we can
  dup it or something), set the correct interface, etc, etc…
• Take all these complicated things out of the servers, to separate process, to
  handle these bad things somewhere else (like asynchronous operations, etc, etc).

While the third one is probably open to discussion, I think we should do the
first two, and the first one of them now.

Thank you

With regards

-- 
Fragile. Do not turn umop ap1sdn!

Michal 'vorner' Vaner
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <https://lists.isc.org/pipermail/bind10-dev/attachments/20120221/34d6ab68/attachment.bin>


More information about the bind10-dev mailing list