Hi,

Jakub wrote:
> On 04/12/2012 09:19 AM, Jiri Svoboda wrote:
> > Hopefully I would be able to merge another part of my work this or next
> > week. In my branch UDP should be already on par with the mainline. Plus
> > the annoying ARP bug is not present.
> 
> I actually still seem to have the problem with ARP. In QEMU:
> 
> # inetcfg create 10.0.2.15/24 net/eth2 v4a
> # ping 10.0.2.2
> # netecho -v -p 8080
> 
> Depending on whether the ping command to the gateway is used or not, the
> nc command from Linux will succeed or not:
> 
> # echo foo | nc -u 127.0.0.1 8080
> 
> Isn't this rather some limitation of the QEMU ARP server?
Well if you are using Qemu port forwarding .. I did not test this at all - I 
only tested with bridged networking. Either it is a bug in Qemu or some 
incompatibility between Qemu ARP and HelenOS ARP.

> Btw, when I kill netecho, I cannot start another instance because it
> complains about the address being in use.
Sounds like the socket API does not release a socket when the client is 
disconnected.

Please file tickets for the problems described above and I will look into them 
when I have some time on my hands.

Thanks
-Jiri

_______________________________________________
HelenOS-devel mailing list
[email protected]
http://lists.modry.cz/cgi-bin/listinfo/helenos-devel

Reply via email to