Eth0 conection stops several times at day

Hello

i have been working no problem since some months ago with my Suse 11.1 64 bits, until a week ago that Udater Aplet asked me to install some new im portant updates and i acepted.

I dont know what did he install but since then the Ethernet conection oges down each 30 minutes or so, geting corrupt so much networks files i have open at that time from the local server.

I must restar Linux for it to work again or do an /etc/rc.d/network restart

Althought files that were open use to get corrupt and i must start doing them again

The message at the log is:

May 27 17:22:56 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.
May 27 17:22:58 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link up.
May 27 17:22:59 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.
May 27 17:23:00 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link up.
May 27 17:23:01 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.

and several ones so.

Plesase i need to repair it as soon as posible or will have to reinstall whole linux because is the computer at work and producing so much problems at it.

regards

guanche

Did you give another networkcard a chance? Please post output of “dmesg | tail -20” from right after disconnect.

A kernel update would have changed the NIC driver, which looks like the Nvidia NIC.

Are you using the Linux drivers or did you install a Nvidia provided driver? I believe Nvidia NICs are handled by the GPLed forcedeth module now. If you did install a Nvidia provided driver you may have to recompile against the current kernel. Then again, I don’t know how the old driver could possibly be loaded with a new kernel since the Nvidia build process would have left the module in the old module directories.

You could also revert the kernel to see if that fixes the problem.

rpm -qa --last | head -50

will tell you the last 50 package updates and when they happened. If you see kernel-something on that list, the kernel was updated. Is there one from a week ago?

Maybe this is related to a bug present, which has been fixed in 2.6.30-rc8, just released now by Linus

Ed Swierk (1):
forcedeth: add phy_power_down parameter, leave phy powered up by default (v2)

Not very likely, as this is a new feature added and not a bugfix, reread the patch description thoroughly.

However, packages with this new feature are on their way (just building) in my repo (home:Akoellh), look for “forcedeth-backport-kmp-$FLAVOR”.

This is one of them:


Jun 2 07:51:03 taginaste ifdown: eth0 device: nVidia Corporation MCP73 Ethernet (rev a2)
Jun 2 07:51:04 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link up.
Jun 2 07:51:04 taginaste avahi-daemon[2990]: Withdrawing address record for 192.200.20.67 on eth0.
Jun 2 07:51:04 taginaste avahi-daemon[2990]: Leaving mDNS multicast group on interface eth0.IPv4 with address 192.200.20.67.
Jun 2 07:51:04 taginaste avahi-daemon[2990]: Interface eth0.IPv4 no longer relevant for mDNS.
Jun 2 07:51:04 taginaste avahi-daemon[2990]: Withdrawing address record for fe80::221:85ff:fe6b:2b91 on eth0.
Jun 2 07:51:05 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.
Jun 2 07:51:05 taginaste ifdown: vboxnet0
Jun 2 07:51:05 taginaste ifdown: No configuration found for vboxnet0
Jun 2 07:51:05 taginaste ifdown: Nevertheless the interface will be shut down.
Jun 2 07:51:06 taginaste SuSEfirewall2: SuSEfirewall2 not active
Jun 2 07:51:07 taginaste ifup: eth0 device: nVidia Corporation MCP73 Ethernet (rev a2)
Jun 2 07:51:07 taginaste kernel: eth0: no link during initialization.
Jun 2 07:51:07 taginaste kernel: ADDRCONF(NETDEV_UP): eth0: link is not ready
Jun 2 07:51:07 taginaste ifup: eth0
Jun 2 07:51:07 taginaste ifup: IP address: 192.200.20.67/24
Jun 2 07:51:07 taginaste ifup:
Jun 2 07:51:07 taginaste avahi-daemon[2990]: Joining mDNS multicast group on interface eth0.IPv4 with address 192.200.20.67.
Jun 2 07:51:07 taginaste avahi-daemon[2990]: New relevant interface eth0.IPv4 for mDNS.
Jun 2 07:51:07 taginaste avahi-daemon[2990]: Registering new address record for 192.200.20.67 on eth0.IPv4.
Jun 2 07:51:07 taginaste SuSEfirewall2: SuSEfirewall2 not active
Jun 2 07:51:07 taginaste ifup: vboxnet0
Jun 2 07:51:07 taginaste ifup: No configuration found for vboxnet0
Jun 2 07:51:07 taginaste SuSEfirewall2: SuSEfirewall2 not active
Jun 2 07:51:08 taginaste kernel: eth0: link up.
Jun 2 07:51:08 taginaste kernel: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Jun 2 07:51:09 taginaste kernel: nfs: server 192.200.20.201 OK
Jun 2 07:51:09 taginaste kernel: nfs: server 192.200.20.201 OK
Jun 2 07:51:09 taginaste kernel: nfs: server 192.200.20.201 OK
Jun 2 07:51:10 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.
Jun 2 07:51:10 taginaste avahi-daemon[2990]: Registering new address record for fe80::221:85ff:fe6b:2b91 on eth0.*.
Jun 2 07:51:11 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link up.
Jun 2 07:51:12 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.
Jun 2 07:51:14 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link up.
Jun 2 07:51:14 taginaste kernel: nv_stop_tx: TransmitterStatus remained busy<6>eth0: link down.

I think i have not installed Nvidia drivers, perhaps i did it for Video Card but not sure.

This is what hardware information say about Network Card:

http://www.guanche.org/fotos/share/tarjeta_red.jpg


The answer to rpm -qa --last | head -50 is:

libsnmp15-5.4.2.1-5.2.1 vie 29 may 2009 08:54:10 WEST
snmp-mibs-5.4.2.1-5.2.1 vie 29 may 2009 08:54:05 WEST
yast2-samba-server-2.17.9-0.1.2 vie 29 may 2009 08:54:02 WEST
cyrus-sasl-plain-2.1.22-182.2.1 jue 28 may 2009 08:44:03 WEST
cyrus-sasl-crammd5-2.1.22-182.2.1 jue 28 may 2009 08:44:03 WEST
cyrus-sasl-digestmd5-2.1.22-182.2.1 jue 28 may 2009 08:44:02 WEST
cyrus-sasl-gssapi-2.1.22-182.2.1 jue 28 may 2009 08:44:01 WEST
cyrus-sasl-2.1.22-182.2.1 jue 28 may 2009 08:44:00 WEST
cyrus-sasl-32bit-2.1.22-182.2.1 jue 28 may 2009 08:43:57 WEST
openssl-certs-0.9.8h-25.2.6 mié 27 may 2009 08:31:50 WEST
openssl-0.9.8h-28.9.1 mié 27 may 2009 08:31:48 WEST
libopenssl0_9_8-32bit-0.9.8h-28.9.1 mié 27 may 2009 08:31:46 WEST
libopenssl0_9_8-0.9.8h-28.9.1 mié 27 may 2009 08:31:42 WEST
kernel-debug-base-2.6.27.21-0.1.2 mié 27 may 2009 08:31:11 WEST
SDL-32bit-1.2.13-104.1 mar 26 may 2009 12:08:23 WEST
pam-devel-1.0.2-13.4 mar 26 may 2009 12:08:22 WEST
flash-plugin-10.0.22.87-release mar 26 may 2009 11:35:24 WEST
opera-9.64-1.1.1 lun 25 may 2009 16:46:15 WEST
inkscape-extensions-dia-0.46-62.33 vie 22 may 2009 13:44:24 WEST
dia-0.96.1-89.106 vie 22 may 2009 13:44:10 WEST
libgnomeprint-2.18.5-1.21 vie 22 may 2009 13:44:00 WEST
dia-lang-0.96.1-89.106 vie 22 may 2009 13:43:58 WEST
libgnomecups-0.2.3-108.12 vie 22 may 2009 13:43:56 WEST
libxine1-32bit-1.1.15-23.4.1 mié 20 may 2009 08:10:54 WEST
yast2-network-2.17.77-0.1.1 mié 20 may 2009 08:10:48 WEST
ntp-4.2.4p6-2.2.1 mié 20 may 2009 08:10:41 WEST
aubio-audacity-plugin-0.3.2-0.pm.3 mar 19 may 2009 09:17:48 WEST
ghostscript-x11-8.62-31.43.1 lun 18 may 2009 08:02:18 WEST
ghostscript-omni-8.62-31.43.1 lun 18 may 2009 08:02:03 WEST
ghostscript-library-8.62-31.43.1 lun 18 may 2009 08:01:25 WEST
libgimpprint-4.2.7-31.43.1 lun 18 may 2009 08:01:03 WEST
ghostscript-fonts-std-8.62-31.43.1 lun 18 may 2009 08:00:49 WEST
ghostscript-fonts-other-8.62-31.43.1 lun 18 may 2009 08:00:42 WEST
ffmpeg-debugsource-0.5-0.pm.3 vie 15 may 2009 12:07:13 WEST
ffmpeg-debuginfo-0.5-0.pm.3 vie 15 may 2009 12:07:12 WEST
flumotion-lang-0.5.4-0.pm.1 vie 15 may 2009 11:57:59 WEST
flumotion-0.5.4-0.pm.1 vie 15 may 2009 11:57:52 WEST
python-dateutil-1.4.1-1.9 vie 15 may 2009 11:57:45 WEST
python-kiwi-1.9.25-0.pm.1 vie 15 may 2009 11:57:43 WEST
python-crypto-2.0.1-28.115.1 vie 15 may 2009 11:57:40 WEST
gstreamer-0_10-plugins-good-extra-0.10.10-3.21.1 vie 15 may 2009 11:57:37 WEST
python-twisted-conch-8.1.0-1.37 vie 15 may 2009 11:57:35 WEST
python-twisted-names-8.0.0-1.58 vie 15 may 2009 11:57:31 WEST
libwmf-0.2.8.4-206.24.4 vie 15 may 2009 11:08:33 WEST
yast2-bootloader-2.17.59-0.1.1 vie 15 may 2009 11:08:30 WEST
gnome-session-2.24.1-6.3.1 jue 14 may 2009 11:00:57 WEST
libgnomeui-32bit-2.24.0-1.36.2 jue 14 may 2009 11:00:53 WEST
libgnomeui-2.24.0-1.36.1 jue 14 may 2009 11:00:51 WEST
MozillaFirefox-translations-3.0.10-1.1.1 mié 13 may 2009 08:46:16 WEST
mozilla-xulrunner190-translations-1.9.0.10-1.1.1 mié 13 may 2009 08:46:10 WEST

Should i recompile kernel?

Thank you so much all for your help.

guanche

Ok, you are using the forcedeth driver, according to the info.

There was an update to kernel-debug-base on 27th May. Identify what kernel you are running (kernel-default, kernel-pae, etc) using uname -a, then do

rpm -qi kernel-whatever

That will tell you when it was installed.

“rpm -qi kernel-whatever” says:

el paquete kernel-whatever no está instalado

(kernel-wahtever is not installed)

Should it be installed?

regards

guanche

replace -whatever with either -default or -pae

the -whatever thing ken_yap mentioned is because he doesn’t know which kernel you have installed

Oh sorry :shame: :shame: :shame:

so brute here.


rpm -qi kernel-default

Name : kernel-default Relocations: (not relocatable)
Version : 2.6.27.21 Vendor: openSUSE
Release : 0.1.2 Build Date: jue 02 abr 2009 00:56:17 WEST
Install Date: mar 14 abr 2009 08:17:26 WEST Build Host: crumb
Group : System/Kernel Source RPM: kernel-default-2.6.27.21-0.1.2.nosrc.rpm
Size : 57898944 License: GPL v2 only
Signature : RSA/8, jue 02 abr 2009 01:01:55 WEST, Key ID b88b2fd43dbdc284
Packager : Submitting Bug Reports - openSUSE
URL : The Linux Kernel Archives
Summary : The Standard Kernel
Description :
The standard kernel for both uniprocessor and multiprocessor systems.

Source Timestamp: 2009-03-31 14:50:44 +0200
GIT Revision: 23c791d70fe628d2d28a4aac869684089bcd3193
GIT Branch: SLE11_BRANCH
Distribution: openSUSE 11.1

Thank you

regards all

guanche

It was installed 14 April, before you said problems started happening.

Maybe you have a NIC hardware problem. Have you checked the cable? Tried a different cable?

Worst comes to the worst, turn off the onboard NIC and install an addon NIC.

It could be that dates because i spent 2 weeks out of office before i notice the fail.

If it is the problem what could i do?

Could it get solver if i choose another kernel as the Zen or anything so?

Thank you so much

regards

guanche

You could reinstall the previous version of the kernel-default package.

How could i do that?

Thank you

regards

guanche

I’ve only ever reverted a package using zypper, not YaST so somebody else will have to confirm if it is possible using YaST. Backup your file just in case it goes wrong.

Is so easy at Yast, have found how to do it… i am going to try to see if it works now.

Thank you

regards

guanche