[e1000e] Intel 82579V suffering from packet drop with e1000e 1.2.20

Bug #756485 reported by Johannes Postler
66
This bug affects 13 people
Affects Status Importance Assigned to Milestone
Fedora
Unknown
Unknown
linux (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

Natty beta 1 currently ships version 1.2.20 of the e1000e driver. It suffers from massive packet drops (66%) and gives a very unstable network connection. The current upstream version1.3.10a fixes the issue, so I suggest including it as soon as possible.

ProblemType: Bug
DistroRelease: Ubuntu 11.04
Package: linux-image-2.6.38-8-generic 2.6.38-8.41 [modified: lib/modules/2.6.38-8-generic/kernel/drivers/net/e1000e/e1000e.ko]
Regression: No
Reproducible: Yes
ProcVersionSignature: Ubuntu 2.6.38-8.41-generic 2.6.38.2
Uname: Linux 2.6.38-8-generic x86_64
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.23.
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: PCH [HDA Intel PCH], device 0: ALC892 Analog [ALC892 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: didi 1293 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'PCH'/'HDA Intel PCH at 0xfe620000 irq 53'
   Mixer name : 'Intel CougarPoint HDMI'
   Components : 'HDA:10ec0892,80862002,00100302 HDA:80862805,80862805,00100000'
   Controls : 36
   Simple ctrls : 19
Date: Sun Apr 10 13:05:34 2011
HibernationDevice: RESUME=UUID=2af43bc5-5187-49ba-8280-e0e37b7c12ce
InstallationMedia: Ubuntu 11.04 "Natty Narwhal" - Beta amd64 (20110330)
IwConfig:
 lo no wireless extensions.

 eth0 no wireless extensions.
ProcEnviron:
 LANGUAGE=de_AT:en
 PATH=(custom, no user)
 LANG=de_AT.UTF-8
 SHELL=/bin/bash
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.38-8-generic root=UUID=19e7faaa-39f1-4991-8eaa-e45e12b9c72f ro quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-2.6.38-8-generic N/A
 linux-backports-modules-2.6.38-8-generic N/A
 linux-firmware 1.50
RfKill:

SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 03/01/2011
dmi.bios.vendor: Intel Corp.
dmi.bios.version: BLH6710H.86A.0105.2011.0301.1654
dmi.board.asset.tag: To be filled by O.E.M.
dmi.board.name: DH67BL
dmi.board.vendor: Intel Corporation
dmi.board.version: AAG10189-204
dmi.chassis.type: 3
dmi.modalias: dmi:bvnIntelCorp.:bvrBLH6710H.86A.0105.2011.0301.1654:bd03/01/2011:svn:pn:pvr:rvnIntelCorporation:rnDH67BL:rvrAAG10189-204:cvn:ct3:cvr:

Revision history for this message
Johannes Postler (josi) wrote :
Brad Figg (brad-figg)
Changed in linux (Ubuntu):
status: New → Confirmed
Revision history for this message
Paul Boven (p-boven) wrote :

Confirming that this is a serious network problem. One of our machines has about 30% of packet loss, the other about 15%.
The packet loss does not occur when the link is at 1Gb/s, but does occur at 100Mb/s (haven't tried 10Mb/s).

lspci:
00:19.0 Ethernet controller: Intel Corporation 82579V Gigabit Network Connection (rev 05)

Workaround: install the up-to-date driver from Intel and then run "update-initramfs -u".

Revision history for this message
Christian Reis (kiko) wrote :

This wasn't fixed for Natty; would be nice to get a testing kernel and eventually SRU to update the driver.

Revision history for this message
gusfm (gusfmora) wrote :

Another workaround is to force it to 10 Mb/s.

sudo ethtool -s eth0 speed 10 autoneg off

Revision history for this message
Ondrej (kocian-on) wrote :
Revision history for this message
hevers (h-evers-kiel-media) wrote :

latest driver install, as mentioned by Ondrej, didnt do the job for me.
Network connection is still loosing hughe amounts of packets.
Throttling it as described by gusfm seems the only way for me right now.

Revision history for this message
Marco Hunsicker (ubuntu-triemax) wrote :

@hevers: Could it be that the problem reappeared after a reboot? I had to exchange the driver in the boot image in order to make the change permanent. The process is described here:

http://ubuntuforums.org/showthread.php?p=9593614

Revision history for this message
Al-Buggie (alonzo-brown) wrote :

I've attempted all the above with no success.

Install driver from Intel
Disable IPv6
I've not rebooted so have updated the boot loader.

But still having packet loss. Any ETA on an patch for this?

Revision history for this message
Josh Stone (cuviper) wrote :

Fedora 16's kernel 3.1 appears to be working properly, as I noted here:
https://bugzilla.redhat.com/show_bug.cgi?id=713315#c19

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi All,

I've built a test kernel for bug 870127. It may help some of you here. If it does, please let me know. Thanks.

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/870127/comments/2

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

I've also built a Natty test kernel for bug 870127 if that is of use to anyone:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/870127/comments/5

Revision history for this message
Terry Burton (terryburton) wrote :

After testing with affected hardware I can confirm that Intel 82579V NICs experience the same issue as the 82579LM reported in bug 870127 which the test kernels resolve.

Thanks.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi Terry,

Thanks for testing and the feedback. For now, I'm going to mark this bug as a duplicate of bug 870127 since you've tested and confirmed the patch there resolves the issue here on Intel 82579V nics as well. Thanks.

Revision history for this message
Emmanuel Stapf [ES] (xumoqkn5d6or9iyw) wrote :

As pointed by the following threads in the ubuntu forums (http://ubuntuforums.org/showthread.php?p=11290188), it is enough to do:

$ sudo rmmod e1000e
$ sudo modprobe e1000e

and you get back the original performance.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.