Bug#222676: follow-up

2007-04-06 Thread Martín Ferrari
Hi Bernd, On 4/6/07, Bernd Eckenfels <[EMAIL PROTECTED]> wrote: On Fri, Apr 06, 2007 at 08:41:00PM -0300, Martín Ferrari wrote: > OK, I've read it, but I still don't get three things: > > - why ifconfig can set promisc mode and then get it right? if you set promisc over the old ioctl interface,

Bug#222676: follow-up

2007-04-06 Thread Bernd Eckenfels
On Fri, Apr 06, 2007 at 08:41:00PM -0300, Martín Ferrari wrote: > OK, I've read it, but I still don't get three things: > > - why ifconfig can set promisc mode and then get it right? if you set promisc over the old ioctl interface, then it is reported by the kernel correctly. > - why iproute doe

Bug#222676: follow-up

2007-04-06 Thread Martín Ferrari
On 4/6/07, Bernd Eckenfels <[EMAIL PROTECTED]> wrote: > Bernd, may be you can give a link to the net-dev discussion as to get > more information about this issue? I googled a bit but I couldn't find > that discussion. http://www.uwsg.iu.edu/hypermail/linux/net/0202.1/0018.html OK, I've read i

Bug#222676: follow-up

2007-04-06 Thread Bernd Eckenfels
On Thu, Apr 05, 2007 at 11:58:12PM -0300, Martín Ferrari wrote: > Hi, I'm trying to help on the triage of old bugs on net-tools. thanks for your help. > Bernd, may be you can give a link to the net-dev discussion as to get > more information about this issue? I googled a bit but I couldn't find >

Bug#222676: follow-up

2007-04-05 Thread Martín Ferrari
Hi, I'm trying to help on the triage of old bugs on net-tools. I tried to reproduce this, but it seems that /sbin/ip suffers the same problem, at least with current etch kernels. If I put the device on promisc mode with tcpdump, neither of them will see it. If I use ip l set eth0 promisc on, it wo