Re: Duplicate Address Detection misfire?

Doesn't my original suggestion still stand... regardless of how this particular problem is fixed? That is: if the sending MAC address is _our_ MAC address, then the address is not duplicate. It seems a simple change (unless the function that processes the packet would have difficulty determining

Re: kern/180775: [bxe] if_bxe driver broken with Broadcom BCM57711 cards

Old Synopsis: if_bxe driver broken with Broadcom BCM57711 cards New Synopsis: [bxe] if_bxe driver broken with Broadcom BCM57711 cards Responsible-Changed-From-To: freebsd-bugs->freebsd-net Responsible-Changed-By: linimon Responsible-Changed-When: Wed Jul 24 00:57:52 UTC 2013 Responsible-Changed-Wh

Re: Duplicate Address Detection misfire?

Ironically, the win 7 host is using an "em" series card. Specifically, an Intel "Gigabit CT". It is configured in vlan mode and the VMWare bridged connection is connected to a VLAN. Regardless, tho, Intel's own drivers are usually not this bad --- and the driver is up-to-date. On Tue, Jul 23,

Re: SO_REUSEPORT: strange kernel balancer behaviour

On 23 July 2013 10:54, trafdev wrote: > Yes, and if you kill this first thread - second thread will start to receive > connections and so on. > That's why I've used one processes-shared acceptor socket which behaves > better (load is balancing > between processes but equality of distribution is fa

Re: SO_REUSEPORT: strange kernel balancer behaviour

Yes, and if you kill this first thread - second thread will start to receive connections and so on. That's why I've used one processes-shared acceptor socket which behaves better (load is balancing between processes but equality of distribution is far from ideal). Btw as per https://lwn.net/Art

Re: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

Hi, So I've upgraded the system with th stable/9 subversion tree, but the same problems occurs with these network cards. IMHO this driver is broken ... Do you know who I should contact to report the issue ? Thanks again. Sébastien On 23.07.2013 15:09, Sébastien RICCIO wrote: Thanks for th

Re: SO_REUSEPORT: strange kernel balancer behaviour

Answering my own email: SO_REUSEPORT on FreeBSD doesn't load balance incoming connections. Test case: * 8 threads * each creates a TCP socket, listening on port 1667, with SO_REUSEPORT * only the first thread ever sees incoming requests. I think this load distribution feature is useful to imple

Re: SO_REUSEPORT: strange kernel balancer behaviour

On 23 July 2013 00:09, trafdev wrote: > It's like shared acceptor FD and N processes: [snip] looks like mine, but I use threads. > Accept conn callback is called in N processes on each connection, only one > wins, > others exit by errno == EAGAIN case. Overhead is almost zero. > Problem is that

Re: Duplicate Address Detection misfire?

Does that mean, then, that the only fix open to some people is to turn off DAD? I have another idea: Require DAD to inspect the sending MAC address. If the sending MAC address is _our_ MAC address, then the packet is not an indication of a duplicate address? On Tue, Jul 23, 2013 at 2:15 AM, Ki

Re: Duplicate Address Detection misfire?

Sorry for the slow reply. We run mainly ESXi (the bare metal version of VMware, not the desktop versions) but I do remember seeing something sort of similar on a desktop version ages ago. Basically we narrowed it down to a Windows driver bug on the host system's ethernet card. It was basically

RE: kern/180430: [ofed] [patch] Bad UDP checksum calc for fragmented packets

The following reply was made to PR kern/180430; it has been noted by GNATS. From: Meny Yossefi To: John Baldwin Cc: "bug-follo...@freebsd.org" Subject: RE: kern/180430: [ofed] [patch] Bad UDP checksum calc for fragmented packets Date: Tue, 23 Jul 2013 14:01:23 + --_000_F2E47A38E4D0B9499D

Re: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

Thanks for the information Adam. Going to update it with svn. Cheers, Sébastien On 23.07.2013 14:55, Adam Vande More wrote: On Tue, Jul 23, 2013 at 7:49 AM, Sébastien RICCIO > wrote: root@filer-01-a:/# freebsd-update -r 9.2-BETA1 upgrade Looking up update.

Re: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

On Tue, Jul 23, 2013 at 7:49 AM, Sébastien RICCIO wrote: > root@filer-01-a:/# freebsd-update -r 9.2-BETA1 upgrade > > Looking up update.FreeBSD.org mirrors... 4 mirrors found. > Fetching metadata signature for 9.1-RELEASE from update4.freebsd.org... > done. > Fetching metadata index... done. > Ins

Re: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

root@filer-01-a:/# freebsd-update -r 9.2-BETA1 upgrade Looking up update.FreeBSD.org mirrors... 4 mirrors found. Fetching metadata signature for 9.1-RELEASE from update4.freebsd.org... done. Fetching metadata index... done. Inspecting system... done. The following components of FreeBSD seem to

Re: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

Hi Steve, Not yet. As the gol was to prepare a system to be production ready, I went with the latest stable... well at least I thought that 9.1 was the latest stable. I'm going to try it. But that would mean that the driver in the 9.2-prerelease kernel has been modified. Going to take a lo

Re: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

Have you tried a more recent version e.g. 9.2-PRERELEASE or 9/stable? Regards Steve - Original Message - From: "Sébastien RICCIO" To: Sent: Tuesday, July 23, 2013 12:28 PM Subject: FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card) Hi freebsd-net! We recently inst

FreeBSD 9.1 and BCM57711 issues (broadcom 10ge ethernet card)

Hi freebsd-net! We recently installed FreeBSD 9.1 64bit on a Dell PowerEdge R510 system in which we have two BCM57711 (for a total of four 10Gbit interfaces.) We're planning to use it as a storage filer using ZFS/NFS. Actually in test, the filer is connected with two 10gigs interfaces to a 1

Re: SO_REUSEPORT: strange kernel balancer behaviour

It's like shared acceptor FD and N processes: Listening proc: bool HttpServer::Listen(unsigned short port, uint listen_backlog) { LOG4CXX_TRACE(kLogger, "Listen"); if ((sockd_acceptor_ = socket(AF_INET, SOCK_STREAM, 0)) == -1) { LOG4CXX_ERROR_ERRNO(kLogger, "socket"); return