Thanks for the input everyone. I went over notes from previous outages of
this kind and we have checked that the port was not disabled in the past,
but I'll make sure we check again next time to be sure.
I'm unable to run ethregs as root due to what I assume is a security
related kernel config (m
We think that maybe you are having link issues since the problem is
intermittent and moving from machine to machine. If it more repeatable on a
since system we could do more. Please check the link at the switch port to see
if it think it has link on that failing system.
Thanks.
Cheers,
John
ethregs would be my choice. ethtool -d also gives some information.
Todd Fujinaka
Software Application Engineer
Networking Division (ND)
Intel Corporation
todd.fujin...@intel.com
(503) 712-4565
-Original Message-
From: Fujinaka, Todd [mailto:todd.fujin...@intel.com]
Sent: Monday, June 08
It is a switch, and I can certainly check the switch next time this
happens. The only thing that makes me think it's local to the box is the
repeated "Link is Up" messages that occur until reboot.
How would I go about getting a dump of the registers if this happens again?
On Mon, Jun 8, 2015 at
It's difficult to troubleshoot a problem that's erratic and hard to replicate.
A dump of the registers and status from the link partner would be a good idea.
I wonder if it isn't something triggered on the other side (is it a switch?)
that sees something odd and disables the port, and a reboot j
I use dkms to build the igb driver after new kernel installs on my fleet of
servers using the following commands after every yum update:
dkms build -m igb -v 5.2.18
dkms install -m igb -v 5.2.18
About once a month, one of my boxes (a different one each time) will stop
responding after this. Noth
