Date: Thu, 16 Jan 2014 09:53:03 -0600 From: Matthew Grooms <mgrooms@shrew.net> To: freebsd-net@freebsd.org Subject: Re: TCP socket handling lo[X], buggy in 9.x? Message-ID: <52D8005F.8070104@shrew.net> In-Reply-To: <20140114100612.GR8472@FreeBSD.org> References: <52D4E633.1060000@shrew.net> <20140114100612.GR8472@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 1/14/2014 4:06 AM, Gleb Smirnoff wrote: > Matthew, > > On Tue, Jan 14, 2014 at 01:24:35AM -0600, Matthew Grooms wrote: > M> It looks like the 9.2 host is sending a reset before sending every PDU. > M> Weird. In any case, this is really easy to re-produce. Anyone have an > M> idea as to why this started happening? I didn't try FreeBSD 10, but I > M> suspect the problem is the same. > > I failed to reproduce that on couple of FreeBSD 11 boxes. > Gleb, Thanks for the sanity check. I upgraded my vSphere ESXi host, and the problem went away. Maybe some esoteric bug with the vswitch. I can't explain it, and I was seeing the issue on two different VMs that were running 9.2-RELEASE-p2. Super weird. Thanks again, -Matthew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52D8005F.8070104>