Network drop out with D54250WYK2

[HKirk1]

Hi Intel and everybody else,

I have recently bought a NUC D54250WYK2. I BIOS updated it to WYLPT10H.86A.0023.2013.1224.0941 and installed Ubuntu Server 12.04. In general everything works great, but there is on thing...

When transferring large files I experience network dropouts from the NUC. It happens with SFTP, WGET, BitTorrent etc. The connection comes back again after aprox 20 sec. also any SSH connection to the NUC will freeze during these 20 seconds. With bittorrent and wget it is not that big of a problem because the connection is restored but with SFTP it is really annoying because I have to do a login again and with large files this dropout happens a lot.

I have not installed a WIFI card, so I am using a wired connection.

I have read about similar problems but with other version of NUC and it is also my understanding that a BIOS upgrade did the trick for those people.

Is there anybody else with this problem and this NUC version? any solutions?

Br

Holger

[Silvia_L_Intel1]

Holger, I would recommend you to test your system on a Windows environment in order to find out if you get the same behavior.

We have only support Windows operating system.

[HKirk1]

Hi Sylvia,

Thanks for your reply, but no thanks:-) Windows is not an option, I see it as a dead end technology with a high price tag.

I will get back to you if I find any solutions.

Have a nice weekend!

Br

Holger

[HKirk1]

Hi Again,

I think the problem is related to the e1000e driver and the I218-V network card used in the D54250WYK2. If i force the network speed to(sudo ethtool -s eth0 speed 100 duplex full) I don't see the problem at all. So it have nothing to do with large files as first mentioned in the original post.

If I don't force it to 100, but just let it autonegotiated to 1000, my syslog get filled up with these if a put heavy load on the network interface(file transfer, streaming, etc.):

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang:

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] TDH <66>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] TDT <6b>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] next_to_use <6b>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] next_to_clean <65>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] buffer_info[next_to_clean]:

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] time_stamp <100098ed9>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] next_to_watch <66>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] jiffies <100099105>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] next_to_watch.status <0>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] MAC Status <40080083>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] PHY Status <796d>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] PHY 1000BASE-T Status <3c00>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] PHY Extended Status <3000>

Feb 17 14:12:23 rastaplace kernel: [ 2806.793214] PCI Status <10>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang:

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] TDH <66>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] TDT <6b>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] next_to_use <6b>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] next_to_clean <65>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] buffer_info[next_to_clean]:

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] time_stamp <100098ed9>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] next_to_watch <66>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] jiffies <1000992f9>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] next_to_watch.status <0>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] MAC Status <40080083>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] PHY Status <796d>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] PHY 1000BASE-T Status <3c00>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] PHY Extended Status <3000>

Feb 17 14:12:25 rastaplace kernel: [ 2808.792571] PCI Status <10>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang:

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] TDH <66>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] TDT <6b>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] next_to_use <6b>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] next_to_clean <65>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] buffer_info[next_to_clean]:

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] time_stamp <100098ed9>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] next_to_watch <66>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] jiffies <1000994ed>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] next_to_watch.status <0>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] MAC Status <40080083>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] PHY Status <796d>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] PHY 1000BASE-T Status <3c00>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] PHY Extended Status <3000>

Feb 17 14:12:27 rastaplace kernel: [ 2810.791752] PCI Status <10>

Feb 17 14:12:27 rastaplace kernel: [ 2810.806448] e1000e 0000:00:19.0 eth0: Reset adapter

Feb 17 14:12:31 rastaplace kernel: [ 2814.579208] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

Right now I am using the Intel recommended driver for e1000e: 2.5.2, but I have also tried with latest driver(3.0.4) with same result. I will try to take the discussion to: http://sourceforge.net/projects/e1000/ Intel Ethernet Drivers and Utilities | Free software downloads at SourceForge.net maybe they can help. It would be nice to get the full speed. I will update my post if I make some new findings.

Br

Holger