When I tried to install OpenSUSE 11.4, my cable was plugged in and the system could not detect the network connection. I used the default network settings except disabling IPV6 in Yast and I am trying to connect to my university LAN.
I then checked /sys/class/net/
eth0/operstate:up
eth0/carrier:1
They indicated the cable was plugged in.
The following is shown in the etc/sysconfig/network ifcfg-eth0 file:
BOOTPROTO=‘dhcp4’
BROADCAST=’’
ETHTOOL_OPTIONS=’’
IPADDR=’’
MTU=’’
NAME=‘NetXtreme BCM5754 Gigabit Ethernet PCI Express’
NETMASK=’’
NETWORK=’’
REMOTE_IPADDR=’’
STARTMODE=‘ifplugd’
USERCONTROL=‘no’
IFPLUGD_PRIORITY=‘20’
Other information might be needed:
xia@linux-5s1r:~/Desktop> ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
inet 127.0.0.2/8 brd 127.255.255.255 scope host secondary lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:1a:a0:e6:ac:46 brd ff:ff:ff:ff:ff:ff
inet6 fe80::21a:a0ff:fee6:ac46/64 scope link
valid_lft forever preferred_lft forever
3: pan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN
link/ether a6:ca:b2:48:4d:b7 brd ff:ff:ff:ff:ff:ff
inet6 fe80::a4ca:b2ff:fe48:4db7/64 scope link
valid_lft forever preferred_lft forever
xia@linux-5s1r:~/Desktop> ip route
127.0.0.0/8 dev lo scope link
xia@linux-5s1r:~/Desktop> ip -s link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
RX: bytes packets errors dropped overrun mcast
40753 500 0 0 0 0
TX: bytess
I then tried to re-install Opensuse 11.3 (32bit) and Opensuse 11.1 (64bit) which is the one my colleague incorrectly installed on our 32 bit machine). The same thing with those two versions.
Any clue what is wrong here? Thanks for your help! I am new to Linux, so I may made stupid mistakes here.
You don’t have an IP address that matters on any NIC. Either your system
is not requesting one or it is not receiving one from a DHCP server. Do
other machines on your network get assigned addresses via DHCP? What do
you have in /var/log/messages from the time you started your system?
Good luck.
On 04/05/2011 11:36 AM, xiazhang3 wrote:
>
> Hi,
>
> When I tried to install OpenSUSE 11.4, my cable was plugged in and the
> system could not detect the network connection. I used the default
> network settings except disabling IPV6 in Yast and I am trying to
> connect to my university LAN.
>
> I then checked /sys/class/net/
> eth0/operstate:up
> eth0/carrier:1
> They indicated the cable was plugged in.
>
> The following is shown in the etc/sysconfig/network ifcfg-eth0 file:
> BOOTPROTO=‘dhcp4’
> BROADCAST=’’
> ETHTOOL_OPTIONS=’’
> IPADDR=’’
> MTU=’’
> NAME=‘NetXtreme BCM5754 Gigabit Ethernet PCI Express’
> NETMASK=’’
> NETWORK=’’
> REMOTE_IPADDR=’’
> STARTMODE=‘ifplugd’
> USERCONTROL=‘no’
> IFPLUGD_PRIORITY=‘20’
> -----------------------------------------------------------------------------------------
> Other information might be needed:
> xia@linux-5s1r:~/Desktop> ip addr
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
> inet 127.0.0.2/8 brd 127.255.255.255 scope host secondary lo
> inet6 ::1/128 scope host
> valid_lft forever preferred_lft forever
> 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP
> qlen 1000
> link/ether 00:1a:a0:e6:ac:46 brd ff:ff:ff:ff:ff:ff
> inet6 fe80::21a:a0ff:fee6:ac46/64 scope link
> valid_lft forever preferred_lft forever
> 3: pan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state
> UNKNOWN
> link/ether a6:ca:b2:48:4d:b7 brd ff:ff:ff:ff:ff:ff
> inet6 fe80::a4ca:b2ff:fe48:4db7/64 scope link
> valid_lft forever preferred_lft forever
> ----------------------------------------------------------------------------------------------
> xia@linux-5s1r:~/Desktop> ip route
> 127.0.0.0/8 dev lo scope link
> ----------------------------------------------------------------------------------------------
> xia@linux-5s1r:~/Desktop> ip -s link
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> RX: bytes packets errors dropped overrun mcast
> 40753 500 0 0 0 0
> TX: bytess
> ----------------------------------------------------------------------------------------------
> I then tried to re-install Opensuse 11.3 (32bit) and Opensuse 11.1
> (64bit) which is the one my colleague incorrectly installed on our 32
> bit machine). The same thing with those two versions.
>
> Any clue what is wrong here? Thanks for your help! I am new to Linux,
> so I may made stupid mistakes here.
>
> Xia
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
I believe it is not receiving one from a DHCP server, since other machines on my network get assigned addresses via DHCP? As for the /var/log/messages, I need to go to my lab tomorrow to have that information.
Is that possible that the network administrator blocked my network access, since I re-installed the system and it was not recognized by their “security system”? However, I also re-installed other machines in the same network several months ago and they worked just fine.
Sure… there could be any number of administrator-side network settings.
Perhaps the DHCP server doesn’t recognize your MAC address and therefore
fails (a LiveCD/LiveDVD on a working system would prove this pretty easily
in most cases). Perhaps you have 801.1x (as I recall) authentication, in
which case there is some kind of authentication for the user in order to
get on the network fully, though I think in that case there is still some
IP address assigned to get that all going (I am not an expert in that area
so ask your admin).
Good luck.
On 04/05/2011 08:36 PM, xiazhang3 wrote:
>
> I believe it is not receiving one from a DHCP server, since other
> machines on my network get assigned addresses via DHCP? As for the
> /var/log/messages, I need to go to my lab tomorrow to have that
> information.
>
> Is that possible that the network administrator blocked my network
> access, since I re-installed the system and it was not recognized by
> their “security system”? However, I also re-installed other machines in
> the same network several months ago and they worked just fine.
>
> Thanks a lot for your reply.
>
> Xia
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
When you say you have disabled ipv6 just what have you changed? I suspect you haven’t really.
Going on my experiences with another problem, see my posts, IPV6 is available from the kernel what ever is done and it’s use is down to the application. Eg I disabled it in the usual ways at the kernel level by adding to the boot appends. No joy. I disabled it locally in firefox and firefox started working. I found yet another disable mechanism and the Konqueror browser started working. That one looks to be intended for global use but YAST is still communicating with IPV6 and as a result DNS isn’t working reliably. I have left an update running for about 11hrs and it’s only 11% done as a results. The browsers can access the repositories with ease and even download the files.
I wont swear to it but the use of the “seemingly global” disable seems to have slowed the browsers down some what. Response times were excellent.
The usual way of disabling ipv6 in kde by the way doesn’t do anything.
I have found one remaining way of disabling it - recompiling the kernel but I don’t think that would be a very good idea.
On your mac’s it’s unusual for these to be logged into servers. What would visitors etc do? Also changing the OS doesn’t change the mac. The protocols must be the same though.
Thanks for John’s comments on IPV6. I disabled it during the installation process. I did notice you cannot disable it in YAST.
Another thing I am thinking is the network card driver. When I typed ‘dmesg | grep eth0’, I got something like “martian source 255.255.255.255 from 134.xx.xxx.xxx (similar to the other machines’ IP addresses). When I typed the same thing on other machines, I got some information like " OUT=MAC=XXXX, SPC=xxx, DST=XXX, LEN=XXX …”. So, how can I know whether the network card driver is working correctly?
As for the var/log/messages, it contains too many lines. So I cannot post all of them here. The last several are the following:
Apr 6 11:42:21 linux-fw9k kernel: ll header: ff:ff:ff:ff:ff:ff:58:b0:35:fc:91:25:08:00
Apr 6 11:42:34 linux-fw9k dhcpcd[2893]: eth0: timed out
Apr 6 11:42:34 linux-fw9k dhcpcd[2893]: eth0: trying to use old lease in `/var/lib/dhcpcd/dhcpcd-eth0.info’
Apr 6 11:42:34 linux-fw9k dhcpcd[2893]: eth0: broadcasting for a lease
Is there anything special in the message file I need to look for?
Disabling IPv6 is possible from Yast under the Network Services section
where you configure the NIC. Even though you are probably using Network
Manager you can still use one section of the Yast screen for Network
Devices to enable/disable IPv6. You must reboot to disable IPv6.
The messages stuff looks like it is asking for an IP address using device
eth0 but if that’s the end then nothing is responding. Are other systems
on the network configured to use DHCP or are they, perhaps,
statically-assigned IP addresses?
Good luck.
On 04/06/2011 11:36 AM, xiazhang3 wrote:
>
> Thanks for John’s comments on IPV6. I disabled it during the
> installation process. I did notice you cannot disable it in YAST.
>
> Another thing I am thinking is the network card driver. When I typed
> ‘dmesg | grep eth0’, I got something like “martian source
> 255.255.255.255 from 134.xx.xxx.xxx (similar to the other machines’ IP
> addresses). When I typed the same thing on other machines, I got some
> information like " OUT=MAC=XXXX, SPC=xxx, DST=XXX, LEN=XXX …”. So, how
> can I know whether the network card driver is working correctly?
>
> As for the var/log/messages, it contains too many lines. So I cannot
> post all of them here. The last several are the following:
>
> Apr 6 11:42:21 linux-fw9k kernel: ll header:
> ff:ff:ff:ff:ff:ff:58:b0:35:fc:91:25:08:00
> Apr 6 11:42:34 linux-fw9k dhcpcd[2893]: eth0: timed out
> Apr 6 11:42:34 linux-fw9k dhcpcd[2893]: eth0: trying to use old lease
> in `/var/lib/dhcpcd/dhcpcd-eth0.info’
> Apr 6 11:42:34 linux-fw9k dhcpcd[2893]: eth0: broadcasting for a lease
>
> Is there anything special in the message file I need to look for?
>
> Thanks again for your help!
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
Disabling IPV6 that way doesn’t stop yast or konqueror from using it. They still use IPV6 and from time to time the dns request does get through. This suggests that the facility is always there in the kernel come what may.
Editing /etc/sysctrl.conf at the obvious place does cause konqueror to start using ipv4 but but yast carries on as usual and as it does get going infrequently the facility must still be available.
There is a kernel compile option relating to ipv6 but who knows what other effects this may have. It may well result in causing problems in other areas.
By the way I did try that one as well - no effect. What seems to have happened is that the ipv6 disabling is no longer a problem for the kernel to handle.
Something’s keeping your DHCP server from giving out IP addresses then.
Ask your admin what could prevent that.
Good luck.
On 04/06/2011 01:36 PM, xiazhang3 wrote:
>
> The other systems were configured using DHCP not static IP.
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
You’ve confusing IPv6 and DNSv6 requests. Disabling IPv6 in YaST unloads
the kernel modules that handle it so there is no more IPv6 traffic that
goes on the wire. That does not, however, prevent applications (FF and
others) from issuing requests for AAAA records via DNS which can still
happen via IPv4, though that doesn’t really help since the resolved IP is,
by definition, not going to work on the system. Disabling both areas make
sense.
Good luck.
On 04/06/2011 04:36 PM, ajohnw wrote:
>
> Disabling IPV6 that way doesn’t stop yast or konqueror from using it.
> They still use IPV6 and from time to time the dns request does get
> through. This suggests that the facility is always there in the kernel
> come what may.
>
> Editing /etc/sysctrl.conf at the obvious place does cause konqueror to
> start using ipv4 but but yast carries on as usual and as it does get
> going infrequently the facility must still be available.
>
> There is a kernel compile option relating to ipv6 but who knows what
> other effects this may have. It may well result in causing problems in
> other areas.
>
> By the way I did try that one as well - no effect. What seems to have
> happened is that the ipv6 disabling is no longer a problem for the
> kernel to handle.
>
> John
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
I wiped out everything and re-installed the system again. Still no luck. We do not have system administrator for Linux machines. So, we have to fix all the problems ourselves.
There is an additional message in the var/log/messages: dhcpcd[9795]: eth0: lease information file `/var/lib/dhcpcd/dhcpcd-eth0.info’ does not exist.
Another weird one:
Apr 6 17:47:04 linux-9bh2 gnomesu-pam-backend: The gnome keyring socket is not owned with the same credentials as the user login: /tmp/keyring-cTjZVj/control
Apr 6 17:47:04 linux-9bh2 gnomesu-pam-backend: gkr-pam: couldn’t unlock the login keyring.
Another thing is I used the check “installation media” option before re-installation, the output said the DVD is broken. However, I used the same one to install other machines several months ago successfully and during the installation process, no error message shown up this time. Is that still able to cause any problem?
I didn’t mean the Linux admin… I mean the network admin who runs the
DHCP servers.
Good luck.
On 04/06/2011 09:06 PM, xiazhang3 wrote:
>
> Hi,
>
> I wiped out everything and re-installed the system again. Still no
> luck. We do not have system administrator for Linux machines. So, we
> have to fix all the problems ourselves.
>
> There is an additional message in the var/log/messages: dhcpcd[9795]:
> eth0: lease information file `/var/lib/dhcpcd/dhcpcd-eth0.info’ does not
> exist.
> Another weird one:
> Apr 6 17:47:04 linux-9bh2 gnomesu-pam-backend: The gnome keyring
> socket is not owned with the same credentials as the user login:
> /tmp/keyring-cTjZVj/control
> Apr 6 17:47:04 linux-9bh2 gnomesu-pam-backend: gkr-pam: couldn’t
> unlock the login keyring.
>
> Another thing is I used the check “installation media” option before
> re-installation, the output said the DVD is broken. However, I used the
> same one to install other machines several months ago successfully and
> during the installation process, no error message shown up this time. Is
> that still able to cause any problem?
>
> Xia
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
Try the following command as well and post all of the output including the
command itself:
sudo /usr/sbin/dhcpcd-test eth0
Good luck.
On 04/06/2011 09:38 PM, ab wrote:
> I didn’t mean the Linux admin… I mean the network admin who runs the
> DHCP servers.
>
> Good luck.
>
>
>
>
>
> On 04/06/2011 09:06 PM, xiazhang3 wrote:
>
>> Hi,
>
>> I wiped out everything and re-installed the system again. Still no
>> luck. We do not have system administrator for Linux machines. So, we
>> have to fix all the problems ourselves.
>
>> There is an additional message in the var/log/messages: dhcpcd[9795]:
>> eth0: lease information file `/var/lib/dhcpcd/dhcpcd-eth0.info’ does not
>> exist.
>> Another weird one:
>> Apr 6 17:47:04 linux-9bh2 gnomesu-pam-backend: The gnome keyring
>> socket is not owned with the same credentials as the user login:
>> /tmp/keyring-cTjZVj/control
>> Apr 6 17:47:04 linux-9bh2 gnomesu-pam-backend: gkr-pam: couldn’t
>> unlock the login keyring.
>
>> Another thing is I used the check “installation media” option before
>> re-installation, the output said the DVD is broken. However, I used the
>> same one to install other machines several months ago successfully and
>> during the installation process, no error message shown up this time. Is
>> that still able to cause any problem?
>
>> Xia
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
I may not have been clear on my last post. Reconfiguring the ethx gives the option
of ipv6 and ipv4 or ipv4 only etc. This has no effect on the behaviour of the apps I
have mentioned. As they still managed to get dns returns occasionally the
ipv6 facility must be still there and active. Prior to this I used the earlier method
of disabling it - the no ipv6 kernel boot load appends. Again no effect. Firfox works
once it’s own configuration is changed, konqueror once the file I mentioned is
changed to disable ipv6 completely. All of these require reboots of course. YAST
takes no notice these changes at all.
Curious thing about all of this is that ipv6 actual data traffic seems to be rapid. Just
the dns causes problems. I think there is a logical reason for this - but have no
interest in following up and understanding it completely - Like many I just want
to use my PC.
Anyway I have filed this as a bug. Critical as I regard a system that can’t use the
web from day 1 as unusable and the cures are confusing in relationship to older
releases. As above the “cures” do not even work in a rather critical area - software
updates. I assume it will be down graded and hardly voted for when really it aught
to be fixed immediately. This sort of thing does not do linux any good at all and
opensuse even less. Google even shows that the problem crops up time and
time again - on other distro’s as well. I have no interest in which aspect of ipv6
is causing the problem either - it’s irrelevant from a users point of view.
Out of interest my network was plugged when I installed onto clean discs. It could
have even found out about this problem itself.
On 04/07/2011 05:06 AM, xiazhang3 wrote:
>
> We do not have system administrator for Linux machines. So, we
> have to fix all the problems ourselves.
do you have MS-Windows machines with no administrators or fire-trucks
that no one knows how to drive, or . . .
–
CAVEAT: http://is.gd/bpoMD [NNTP via openSUSE 11.3 + KDE4.5.5 +
Thunderbird3.1.8] Can you believe it? This guy Ralph wins $181 million
in the lottery last Wednesday, and then finds the love of his life just
2 days later. Talk about LUCK!