CentOS - Jan 2015 - tg3 network link unstable

Full updated EL6: Does someone have the same behaviour, unstable eth interface
(tg3/no dhcp)?

Jan 19 18:01:46 ane kernel: tg3 0000:04:00.0: eth0: Link is up at 1000 Mbps,
full duplex
Jan 19 18:01:46 ane kernel: tg3 0000:04:00.0: eth0: Flow control is on for TX
and on for RX
Jan 19 18:01:46 ane kernel: tg3 0000:04:00.0: eth0: EEE is enabled
Jan 19 18:01:46 ane NetworkManager[1735]: <info> (eth0): carrier now ON
(device state 8)
Jan 19 18:02:17 ane kernel: tg3 0000:04:00.0: eth0: Link is down
Jan 19 18:02:17 ane NetworkManager[1735]: <info> (eth0): carrier now OFF
(device state 8, deferring action for 4 seconds)
Jan 19 18:02:20 ane kernel: tg3 0000:04:00.0: eth0: Link is up at 1000 Mbps,
full duplex
Jan 19 18:02:20 ane kernel: tg3 0000:04:00.0: eth0: Flow control is on for TX
and on for RX
Jan 19 18:02:20 ane kernel: tg3 0000:04:00.0: eth0: EEE is enabled
Jan 19 18:02:20 ane NetworkManager[1735]: <info> (eth0): carrier now ON
(device state 8)
Jan 19 18:02:23 ane kernel: tg3 0000:04:00.0: eth0: Link is down
Jan 19 18:02:23 ane NetworkManager[1735]: <info> (eth0): carrier now OFF
(device state 8, deferring action for 4 seconds)
Jan 19 18:02:26 ane kernel: tg3 0000:04:00.0: eth0: Link is up at 1000 Mbps,
full duplex
Jan 19 18:02:26 ane kernel: tg3 0000:04:00.0: eth0: Flow control is on for TX
and on for RX
Jan 19 18:02:26 ane kernel: tg3 0000:04:00.0: eth0: EEE is enabled
Jan 19 18:02:26 ane NetworkManager[1735]: <info> (eth0): carrier now ON
(device state 8)
Jan 19 18:02:50 ane kernel: tg3 0000:04:00.0: eth0: Link is down
Jan 19 18:02:50 ane NetworkManager[1735]: <info> (eth0): carrier now OFF
(device state 8, deferring action for 4 seconds)
Jan 19 18:02:53 ane kernel: tg3 0000:04:00.0: eth0: Link is up at 1000 Mbps,
full duplex
Jan 19 18:02:53 ane kernel: tg3 0000:04:00.0: eth0: Flow control is on for TX
and on for RX
Jan 19 18:02:53 ane kernel: tg3 0000:04:00.0: eth0: EEE is enabled
Jan 19 18:02:53 ane NetworkManager[1735]: <info> (eth0): carrier now ON
(device state 8)
Jan 19 18:03:11 ane kernel: tg3 0000:04:00.0: eth0: Link is down
Jan 19 18:03:11 ane NetworkManager[1735]: <info> (eth0): carrier now OFF
(device state 8, deferring action for 4 seconds)
Jan 19 18:03:14 ane kernel: tg3 0000:04:00.0: eth0: Link is up at 1000 Mbps,
full duplex
Jan 19 18:03:14 ane kernel: tg3 0000:04:00.0: eth0: Flow control is on for TX
and on for RX
Jan 19 18:03:14 ane kernel: tg3 0000:04:00.0: eth0: EEE is enabled
Jan 19 18:03:14 ane NetworkManager[1735]: <info> (eth0): carrier now ON
(device state 8)

On Mon, Jan 19, 2015 at 9:10 AM, Leon Fauster
<leonfauster at googlemail.com> wrote:
> Full updated EL6: Does someone have the same behaviour, unstable eth
interface (tg3/no dhcp)?
>
> Jan 19 18:01:46 ane kernel: tg3 0000:04:00.0: eth0: Link is up at 1000
Mbps, full duplex
> Jan 19 18:01:46 ane kernel: tg3 0000:04:00.0: eth0: Flow control is on for
TX and on for RX
You were probably hit a bug described in this upstream article:

https://access.redhat.com/solutions/1302043

Check to see if the tg3 version is 3.137 (modinfo tg3). If so, one
workaround is to downgrade the tg3 driver.  Luckily, ELRepo's kmod-tg3
for EL6 is at 3.136. Install the kmod and see if this resolves the
issue.

Akemi