Date: Tue, 10 Jul 2007 14:37:00 -0400 From: Michael C Voorhis <mvoorhis@cs.wpi.edu> To: "Kevin Gerry" <sfpoof@gmail.com> Cc: Michael C Voorhis <mvoorhis@cs.wpi.edu>, freebsd-current@freebsd.org, Steve Kargl <sgk@troutmask.apl.washington.edu> Subject: Re: Issues with 'xl0' keeping link (bge0, em0 as well?) Message-ID: <18067.53708.882974.602732@cs.wpi.edu> In-Reply-To: <c027a39a0707081230v51099908hae3961af656ace7a@mail.gmail.com> References: <c027a39a0706251819y157a352fu292dc2533c6920b1@mail.gmail.com> <20070626022444.GA6571@troutmask.apl.washington.edu> <18049.19912.746062.704028@cs.wpi.edu> <c027a39a0707081230v51099908hae3961af656ace7a@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Kevin Gerry writes:
> Did you ever have a result to this issue? What's interesting is my em0 NIC
> doesn't seem to have the up/down issue that the xl NICs have.
For me, the "em0 watchdog timeout" issue on my SMP machine seems to be
tied to the onboard em device, which happens to be a PCI-Express
device.
I plugged in another Intel gigabit card, a PCI-X card (an Intel
82545GM "Pro/1000 MT Server Adapter"), and all my problems have
disappeared. I've got my test case running in a while(1) loop, and I
haven't had a watchdog reset or an NFS timeout in several hours.
As far as I can tell, both cards are supported by the em driver in
FreeBSD. I'll be shutting off the PCI-express devices in BIOS and
using the PCI-X card.
Mike.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?18067.53708.882974.602732>