Date: Fri, 25 Nov 2011 03:26:03 +0200 From: Mihai-Catalin Salgau <csalgau-br@bitdefender.com> To: freebsd-net@freebsd.org Subject: Re: vlan limits on e1000? Message-ID: <4ECEEEAB.9070907@bitdefender.com> In-Reply-To: <07255796.20101207031807@bitdefender.com> References: <07255796.20101207031807@bitdefender.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12/7/2010 3:18 AM, Mihai-Catalin Salgau wrote: > Hello Freebsd-net, > > I have two dual port NICs, one Broadcom(bce0,bce1) and one Intel(em0,em1), on FreeBSD 8-stable > (about two weeks old) with a DHCP server running. > I've been successfully using a large number of vlans over bce1,em0 and em1 with iSCSI, > but wanted to switch to AoE(ata over ethernet). I've set vlandevs by round-robin, and got > vlan1 on bce0, vlan2 on em0, vlan3 on em1, vlan4 on bce0....vlan12 on em1. I've binded > net/vblade instances to each interface, but the problem I'm facing now is that while > vlans 1-10 are working properly, vlans 11 and 12 won't see any traffic unless the interface is > in promiscuous mode. I noticed that while trying to attach tcpdump and saw the thing instantly work. > I've had no problems with iSCSI over the same setup, and dhcp packets are getting trough properly. > I've moved those last two vlans to bce0 and they work ok, but I'm a bit locked on why this is happening. > Are there any known limitations on vlans on e1000? > Hi I left this dead for over a year now due to lack of time and the need to have the box up. Sorry about that. As a workaround I moved most of the vlans on the Broadcom interface and left only four on the Intel, as that worked. Now I'm really trying to debug this, so here's what I've done up until now. I've exchanged the Intel cards three times - two other Intels and a Broadcom card and the same thing happened(or didn't. depending on the point of view) So I moved to tcpdumps. All of the following was done on the server, not the clients. While running tcpdump -p -i vlan12 I see 1 AoE request frame received by the server and 1 reply being sent out to the client but that's it. Running without -p(switching to promiscuous mode) sees subsequent requests and replies and everything works right. Doing this on em0 has the same effect, but frames are tagged(duh). So I started with the hacks. I modified vblade(the AoE target I'm using) to read and write 802.1q tagged frames for the vlan I'm testing with and set it on em0. This turned out to work just fine (without promiscuous mode, that is), so I'm currently suspecting BPF as the culprit, as vblade uses BPF to read and write network packets. I'll have a mirror port set on the switch later today so I can see if any packets actually go out when tcpdump says they do, but I'm currently seeing two problems. -tcpdump not showing the actual frames being sent -BPF on VLANs not working properly I'm thinking they might get truncated after pcap and are invalid when they reach the client Any feedback or at least confirmation is welcome. I'm currently running 8-STABLE from about a week ago. Unfortunatelly I can't test CURRENT or 9-STABLE right now.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4ECEEEAB.9070907>