I have no broadband access on a Thinkpad T430 running Leap 15.0 with a KDE/Plasma desktop, Do I need to install (k)networkmanager. I use Ethernet to a hub and to an AT&T modem. Light in the TPs RJ socket is lit when Ethernet plug is inserted and lights on the hub’s RJ socket are lit, but not flashsing. The light of the Ethernet cable from the modem to the hub is lit and flashing. It appears that I don’t have firewall installed. Broadband was working and perhaps installation of xorg-x11-driver-video may be associated with loss of broadband. Wireless on the T430 is turned off.
Leap 15 - KDE/Plasma
I have loaded Ubuntu 18 from a live DVD and found that I have a broadband connection, so this problem must be in the way that I have Leap 15 configured and not a problem with my broadband service.
Wireless is disabled in BIOS
From the popup from the Leap 15 Networks icon in the task bar:
Wired connection 1
Automatically connect to this network when it is available
All users may connect to this network
Restrict to device: eth0 (28:2:44:18:BO:CC)
Duplex: Half
(Other fields are blank)
Use 802.1x security for this connection - Not checked
IPv4 and IPv6 are both set to Automatic and IPv6 Privacy: Default
Hope that helps.
Both NetworkManager and the KDE applet should be installed already. They are a part of a standard KDE install.
You might need to use:
Yast –> System –> Network settings
to switch to using NetworkManager instead of “wicked”. However, I would have expected that to be the default setting already.
The half duplex reported is a bit strange. I would normally expect full duplex, and this might be indicative of a faulty cable, port, or driver.
Some definitive information would be useful here…
Network hardware details
/sbin/lspci -nnk|grep -iA3 net
Query ethernet negotiation and hardware details
/usr/sbin/ethtool eth0
IP address assigned (if any)
ip a
Route(s) (if any)
ip r
I suggest copy/pasting the output to a text file, transfer to a memory stick, and then post the info via a machine with internet connectivity.
I thought it was unlikely that cables or external ports were the problem, but in case I swapped cables with the Thinkpad I am using here running 13.2 and this one still works while the T430 still doesn’t. I also think it is unlikely that the ethernet port on the T430 is faulty since it worked with the live ubuntu boot. Both Thinkpads are connected to a Netgear hub. Drivers might be involved since I had broadband connectivity on the T430 before installing the video driver – xorg-x11-driver-video**.** I have also run these cmds getting the video problems sorted:
mkinitrd
mkinitrf -A
…although I didn’t think to save the responses, though both succeeded.
Here is the trace of the commands you suggested, run as root:
Password:
linux-muav:~ # /sbin/lspci -nnk|grep -iA3 net
00:19.0 Ethernet controller [0200]: Intel Corporation 82579LM Gigabit Network Connection (Lewisville) [8086:1502] (rev 04)
Subsystem: Lenovo Device [17aa:21f3]
Kernel driver in use: e1000e
Kernel modules: e1000e
--
03:00.0 Network controller [0280]: Intel Corporation Centrino Advanced-N 6205 [Taylor Peak] [8086:0085] (rev 34)
Subsystem: Intel Corporation Centrino Advanced-N 6205 (802.11a/b/g/n) [8086:1311]
Kernel driver in use: iwlwifi
Kernel modules: iwlwifi
linux-muav:~ # /usr/sbin/ethtool eth0
Settings for eth0:
Supported ports: TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supported pause frame use: No
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Speed: 10Mb/s
Duplex: Half
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
MDI-X: off (auto)
Supports Wake-on: pumbg
Wake-on: d
Current message level: 0x00000007 (7)
drv probe link
Link detected: yes
linux-muav:~ # ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 28:d2:44:18:b0:cc brd ff:ff:ff:ff:ff:ff
inet 192.168.1.67/24 brd 192.168.1.255 scope global noprefixroute dynamic eth0
valid_lft 85840sec preferred_lft 85840sec
inet6 2600:1702:30b0:b710:692c:d412:4988:5f67/64 scope global temporary dynamic
valid_lft 3537sec preferred_lft 3537sec
inet6 2600:1702:30b0:b710:774b:c057:10b0:3882/64 scope global mngtmpaddr noprefixroute dynamic
valid_lft 3537sec preferred_lft 3537sec
inet6 fe80::50bb:892b:15fd:ee50/64 scope link noprefixroute
valid_lft forever preferred_lft forever
3: wlan0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN group default qlen 1000
link/ether de:c2:a8:31:94:80 brd ff:ff:ff:ff:ff:ff
linux-muav:~ # ip r
default via 192.168.1.254 dev eth0 proto dhcp metric 20100
192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.67 metric 100
linux-muav:~ #
Would it have been preferable here to have reported these as separate code blocks?
While in yast I noticed that the firewall is apparently installed, but not configured. I checked to see if Yast was controlling networks and it was not; the install default is apparently networkmanager.
The ethtool output confirms a problem with negotiation (10Mbps, half-duplex), but it appears that the NIC is getting a DHCP assigned IP address and a valid default route exists.
FWIW, I recall a similar thread a few months back where a user had an unintended NetworkManager connection configuration where the negotiation was set incorrectly. (Usually, no explicit configuration is required.) Anyway, you can check that via the connection editor or examine the connection profile located in the
/etc/NetworkManager/system-connections/ directory.
I think this was the thread…
Yes it seems to be a similar problem, though there a working connection existed, where here NetworkManager thinks there is a connection, where there actually isn’t:
Re: Slow internet on Tumbleweed compared to Fedora
https://forums.opensuse.org/images/misc/quote_icon.png Originally Posted by equeim https://forums.opensuse.org/images/buttons/viewpost-right.png](https://forums.opensuse.org/showthread.php?p=2880621#post2880621)
Yep, after reboot it’s ok too. Connection config file in /etc/NetworkManager/system-connections/ had fields that disabled auto-negotiation and enabled half duplex. No idea how they got there (maybe when I switched from wicked to NetworkManager). I just removed this file and NetworkManager set up everything correctly.
I was a little hesitant to delete an existing configuration file without a backup, so I tried this:
linux-muav:/etc/NetworkManager/system-connections # cp Wired connection 1 /home/n_nullo/Buffer/
cp: cannot stat 'Wired': No such file or directory
cp: cannot stat 'connection': No such file or directory
cp: cannot stat '1': No such file or directory
linux-muav:/etc/NetworkManager/system-connections # ls
Wired connection 1
linux-muav:/etc/NetworkManager/system-connections #
I am not familiar with the Kate editor or with file naming rules in Leap, but since the OS named this connection file, it seems odd that it objects to an attempt to copy that file to another directory. Maybe I am not understanding the error msg “cannot stat <file name>”.
That needs to be:
cp "Wired connection 1" /home/n_nullo/Buffer/
A space is a word separator. So you need quotes to treat that name as a single word.
Yes, connection files copy without a problem. I’ve done that here, to make a backup. However, normally that file does not exist unless you have modified the connection by changing from some of the defaults.
Just curious. I remember a file system rule in the Windows/OS2 world that disallowed spaces in file names, but then I think that got relaxed. That seems to be still a safe rule for Linux file names but not always followed?
**Re: Slow internet on Tumbleweed compared to Fedora** https://forums.opensuse.org/images/misc/quote_icon.png Originally Posted by **equeim** https://forums.opensuse.org/images/buttons/viewpost-right.png](https://forums.opensuse.org/showthread.php?p=2880621#post2880621) Yep, after reboot it's ok too. Connection config file in /etc/NetworkManager/system-connections/ had fields that disabled auto-negotiation and enabled half duplex. No idea how they got there (maybe when I switched from wicked to NetworkManager). I just removed this file and NetworkManager set up everything correctly.
I copied then deleted the “Wired connection 1” file, rebooted and that file got recreated, still with Allow auto-negotiation initially not enabled. I clicked the enable box, rebooted and the Allow auto-negotiation was still enabled, but no broadband connection.
I think the screens I am seeing in the Edit your Network Connections frame may be slightly different than before. For example, I think it was on the then second tab that I saw this:
Restrict to device: eth0 (28:2:44:18:BO:CC)
Duplex: Half
(Other fields are blank)
…and now the Wired tab has no Duplex information, only the Restrict to device, Cloned MAC address and MTU fields plus the Allow auto-negotiation switch.
I have rerun the diagnostic cmds that you sent earlier in the chance that some things have changed.
Password:
linux-muav:~ # /sbin/lspci -nnk|grep -iA3 net
00:19.0 Ethernet controller [0200]: Intel Corporation 82579LM Gigabit Network Connection (Lewisville) [8086:1502] (rev 04)
Subsystem: Lenovo Device [17aa:21f3]
Kernel driver in use: e1000e
Kernel modules: e1000e
--
03:00.0 Network controller [0280]: Intel Corporation Centrino Advanced-N 6205 [Taylor Peak] [8086:0085] (rev 34)
Subsystem: Intel Corporation Centrino Advanced-N 6205 (802.11a/b/g/n) [8086:1311]
Kernel driver in use: iwlwifi
Kernel modules: iwlwifi
linux-muav:~ # /usr/sbin/ethtool eth0
Settings for eth0:
Supported ports: TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supported pause frame use: No
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Speed: 10Mb/s
Duplex: Half
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
MDI-X: off (auto)
Supports Wake-on: pumbg
Wake-on: d
Current message level: 0x00000007 (7)
drv probe link
Link detected: yes
linux-muav:~ # ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 28:d2:44:18:b0:cc brd ff:ff:ff:ff:ff:ff
inet 192.168.1.67/24 brd 192.168.1.255 scope global noprefixroute dynamic eth0
valid_lft 83672sec preferred_lft 83672sec
inet6 2600:1702:30b0:b710:b8fc:f57b:e712:4f53/64 scope global temporary dynamic
valid_lft 3512sec preferred_lft 3512sec
inet6 2600:1702:30b0:b710:774b:c057:10b0:3882/64 scope global mngtmpaddr noprefixroute dynamic
valid_lft 3512sec preferred_lft 3512sec
inet6 fe80::50bb:892b:15fd:ee50/64 scope link noprefixroute
valid_lft forever preferred_lft forever
3: wlan0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN group default qlen 1000
link/ether 9e:4f:d0:c7:cc:cb brd ff:ff:ff:ff:ff:ff
linux-muav:~ # ip r
default via 192.168.1.254 dev eth0 proto dhcp metric 20100
192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.67 metric 100
linux-muav:~ #
I’ve also done screenshots of the Network Settings screen:
…Well I thought I could, but the Insert Image tool wants a url and doesn’t allow me to browse local volumes for a file. If there is a way to refer to a local file as a url, I’d be glad to learn how.
I think I am here:
- Hardware/cables have been proven good
- NetworkManager has control of network parameters
- NetworkManager thinks I have a network connection, but I don’t
- DCHP has assigned an address
- According to the LEDs on my hub, the port to the TP 430 is at least getting polled
- Deleting and allowing the connection file to be recreated doesn’t solve the problem
You are allowed to use spaces in filenames, but spaces are used by commands to delimit options etc, so you need to enclose such filenames within quotes, or use "\ " eg
cat Wired\ connection\ 1
https://www.hecticgeek.com/2014/02/spaces-file-names-command-line/
I have rerun the diagnostic cmds that you sent earlier in the chance that some things have changed.
Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Half
Auto-negotiation is failing for some reason.
I’ve also done screenshots of the Network Settings screen:
…Well I thought I could, but the Insert Image tool wants a url and doesn’t allow me to browse local volumes for a file. If there is a way to refer to a local file as a url, I’d be glad to learn how.
Use https://susepaste.org/
In the top-right you can select ‘Image’ and upload your local image to it. Set the ‘Delete After’ period to something sensible. Don’t copy the URL that it provides (as it’s missing the file suffix), but rather right-click on the image presented and select ‘Copy Image Location’, and use that complete URL in your post.
I think I am here:
- Hardware/cables have been proven good
- NetworkManager has control of network parameters
- NetworkManager thinks I have a network connection, but I don’t
- DCHP has assigned an address
- According to the LEDs on my hub, the port to the TP 430 is at least getting polled
- Deleting and allowing the connection file to be recreated doesn’t solve the problem
Yes, agreed.
[QUOTE] I think I am here:
- Hardware/cables have been proven good
- NetworkManager has control of network parameters
- NetworkManager thinks I have a network connection, but I don’t
- DCHP has assigned an address
- According to the LEDs on my hub, the port to the TP 430 is at least getting polled
- Deleting and allowing the connection file to be recreated doesn’t solve the problem
Yes, agreed.
[/QUOTE]
I started this project because I was running openSUSE 13.1–apparently past EOL – on a 15 year old Thinkpad which tend to run forever, but there exceptions. I bought a high-end, highly rated used Thinkpad, a mate to the one I have. I committed to Leap15 because of a long history with openSUSE Linux which seems like one of the most professional distributions. I’ve had very excellent and responsive technical support from this forum. However, if I can’t solve this problem with networks support and can’t connect to the internet, the project will be a failure. This is in no way a complaint; only a statement of reality and a plea.
I get that. Are you able to provide the following dmesg output please?
dmesg|grep e1000e
It might be worth trying the current stable kernel perhaps (but be aware if using proprietary graphics drivers)…
https://forums.opensuse.org/showthread.php/531974-Network-connection-Problems-(Hardware-Driver-)-Opensuse-Leap-15?p=2872444#post2872444
Here’s a couple of interesting archlinux threads where it was found that Windows WoL/network power management settings were impacting on the NIC when booting into Linux…
https://bbs.archlinux.org/viewtopic.php?id=191981
https://bbs.archlinux.org/viewtopic.php?id=189633
Of course, only relevant if dual booting with Windows on this laptop.
Thought I had done this…
Password:
linux-muav:~ # dmesglgrep e1000e
If 'dmesglgrep' is not a typo you can use command-not-found to lookup the package that contains it, like this:
cnf dmesglgrep
linux-muav:~ # cnf dmesglgrep e1000e
dmesglgrep: searching ...linux-muav:~ #
I swapped the drives back to their original T430s (and MS is so happy). The Leap machine is no happier and based on my original video problems after the initial swap, I tried to run, hoping it might fix the network problem:
Password:
linux-muav:~ # mkinitrd -A
Creating initrd: /boot/initrd-4.12.14-lp150.11-default
dracut: Executing: /usr/bin/dracut --no-host-only --logfile /var/log/YaST2/mkinitrd.log --force /boot/initrd-4.12.14-lp150.11-default 4.12.14-lp150.11-default
dracut: No permission to write to /boot.
update-bootloader: 2018-12-31 09:17:16 <3> update-bootloader-2729 run_command.274: '/usr/lib/bootloader/grub2/config' failed with exit code 1, output:
<<<<<<<<<<<<<<<<
+ /usr/sbin/grub2-mkconfig -o /boot/grub2/grub.cfg
/usr/sbin/grub2-mkconfig: line 277: /boot/grub2/grub.cfg.new: Read-only file system
>>>>>>>>>>>>>>>>
Updating bootloader failed
linux-muav:~ #
I am running this as root, and as such I don’t remember having a permissions problem before.
Be careful with your syntax. It’s not the letter ‘l’ - the ‘|’ (pipe) symbol takes the output from one command and uses it as the input for the next command.
dmesg | grep...
When I was deciding whether to install Tumbleweed or Leap, I understood Leap to be “the most usable Linux distribution and stabilized operating system.” I assumed, maybe wrongly, that Leap 15 would be using the most stable version of the kernel.
When I installed Leap15, I assume it contained a recent version of the kernal and some “generic” or non-proprietary drivers that were chosen by the installer as best fit for my hardware. Best fit wasn’t good enough and produced significant problems in the task bar and vertical control. I think with the plain-vanilla state, I still had broadband support. I suspect that installing the different video driver caused the problem with NetworkManager. Installing a “current stable kernel” , which I assume does not uninstall the updated video driver and I run the risk that there will be incompatibility. That sounds like a risky choice for someone who does not know how to backout a kernel update.
Would I be better off running Leap 42.xx
BTW I am not dual booting. safer to have a separate machine for each OS.
No, it is not using a recent kernel. It is essentially built using the same kernel/core as its contemporary SLED counterpart…
https://en.opensuse.org/Portal:Leap/Leap_kernel_version
Best fit wasn’t good enough and produced significant problems in the task bar and vertical control. I think with the plain-vanilla state, I still had broadband support. I suspect that installing the different video driver caused the problem with NetworkManager.
No, the graphics drivers and network driver behaviors are entirely unrelated.
Installing a “current stable kernel” , which I assume does not uninstall the updated video driver and I run the risk that there will be incompatibility. That sounds like a risky choice for someone who does not know how to backout a kernel update.
If you’ve installed a proprietary graphics driver, then yes you would need to reinstall the driver (built with respect to the current running kernel). If you don’t have the knowledge and/or time to do this, then don’t install this kernel.
Would I be better off running Leap 42.xx
Simple answer - I don’t know. It would really depend on whether the kernel used in Leap 15 is suffering from some kind of regression, or whether your laptop’s particular NIC chipset would benefit from a later version of the driver. We haven’t really identified the underlying cause of the negotiation/connectivity issue.
BTW I am not dual booting. safer to have a separate machine for each OS.
Ok, but in the context of the previous posts this is irrelevant. If you’ve recently booted Windows OS and the network power management settings described previously are employed, then it is possible that the NIC has been left in a state which will effectively leave it crippled in the Linux environment. This was speculation, and some experimentation may be required (as mentioned in the pages linked to already) to check that this is not the reason for the connectivity issue you’re encountering.
Don’t forget to post the requested dmesg output. Pay close attention to syntax.