e1000 or e1000e or ixgbe interface has a link down/up event when changing the ring settings with 'ethtool -G' in RHEL 6

Solution Unverified - Updated -

Issue

  • While changing the ring buffer on one of our Interface which was in bond (Active -backup mode); we saw that primary and slave interfaces got interchanged
  • Changing the ring buffer settings for e1000 or e1000e or ixgbe interfaces causes a link loss
  • When running ethtool -G eth5 rx 2048 the bond reported the slave as having gone down and back up quickly:
Jul 20 00:15:43 example kernel: bonding: bond0: link status definitely down for interface eth4, disabling it
Jul 20 00:15:43 example kernel: bonding: bond0: making interface eth0 the new active one.
Jul 20 00:15:46 example kernel: e1000e: eth4 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
Jul 20 00:15:46 example kernel: bond0: link status definitely up for interface eth4, 1000 Mbps full duplex.
Jul 20 00:15:47 example kernel: e1000e: eth4 NIC Link is Down
Jul 20 00:15:47 example kernel: bonding: bond0: link status definitely down for interface eth4, disabling it
Jul 20 00:15:49 example kernel: e1000e: eth4 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
Jul 20 00:15:49 example kernel: bond0: link status definitely up for interface eth4, 1000 Mbps full duplex.
Jul 20 00:34:52 example kernel: bonding: bond1: link status definitely down for interface eth5, disabling it
Jul 20 00:34:52 example kernel: bonding: bond1: making interface eth1 the new active one.
Jul 20 00:34:55 example kernel: e1000e: eth5 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
Jul 20 00:34:55 example kernel: bond1: link status definitely up for interface eth5, 1000 Mbps full duplex.
Oct 19 12:22:10 kernel: bonding: bond0: link status definitely down for interface eth9, disabling it  
Oct 19 12:22:10 kernel: ixgbe 0000:47:00.1: eth9: detected SFP+: 5
Oct 19 12:22:11 kernel: bonding: bond0: link status definitely down for interface eth10, disabling it  
Oct 19 12:22:11 kernel: ixgbe 0000:04:00.1: eth10: detected SFP+: 5
Oct 19 12:22:12 kernel: ixgbe 0000:47:00.1: eth9: NIC Link is Up 10 Gbps, Flow Control: None
Oct 19 12:22:12 kernel: bond0: link status definitely up for interface eth9, 10000 Mbps full duplex.
Oct 19 12:22:13 kernel: ixgbe 0000:04:00.1: eth10: NIC Link is Up 10 Gbps, Flow Control: None

Environment

  • Red Hat Enterprise Linux (RHEL) 6
  • ethtool
  • Interface using e1000 or e1000e or ixgbe

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content