[PD] Netreceive + osc issue
Roman Haefeli
reduzent at gmail.com
Wed Mar 24 22:20:11 CET 2021
On Wed, 2021-03-24 at 21:41 +0100, IOhannes m zmölnig wrote:
> On 3/24/21 21:02, Rick Snow wrote:
> > Hi all,
> >
> > I wonder if anyone can help me out with this issue.
> >
> > I’m receiving an OSC message using [netreceive -u -b 7002] from
> > another piece of software on the same computer. Netreceive is
> > producing an error in the console reading:
> >
> > “recv (bin): a message sent on a datagram socket was larger than
> > the internal message buffer or some other network limit, or the
> > buffer used to receive a datagram into was smaller than the
> > datagram itself. (10040)”
> >
>
> there has been discussion at [903] about enlarging the network buffer
> to
> the maximum possible frame size for UDP packages.
Alternatively, you could use [udpreceive] from iemnet that doesn't have
this limit set.
>
> right now that size is rather small (4096 bytes), which is usually
> enough for real networked traffic (as most networking components
> will
> impose a default maximum package size of 1500 bytes),
Datagrams larger than the MTU are split into smaller segments fitting
into the MTU. On the receiving end the segments are re-assembled to
restore the original datagram (I'm not sure which but I believe this is
done by a lower layer of the network stack). It's still not advised to
use larger datagrams than the MTU because if any of the segments is
lost the whole datagram is lost.
> but when the
> computer is "networking" with itself (communicating via "localhost")
> that limit can be easily blown.
>
> until the corresponding PR [1122] is accepted (and there is a bit of
> reluctance accepting it),
That's actually sad. I'm in favor of the maximum size supported by
UDP.
Roman
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: This is a digitally signed message part
URL: <http://lists.puredata.info/pipermail/pd-list/attachments/20210324/ee0b793d/attachment.sig>
More information about the Pd-list
mailing list