Need some help with network setup and internet access SUSE 13.2

English Network/Internet
1

Hey guys,

couple questions about the network services with SUSE 13.2. My understanding is that there’s wicked, network service, and NetworkManager and the three of these don’t play nicely with each other … after looking around on various threads I think I may have done more damage than good on trying to get my computer to connect to the internet. The background is that i’m trying to get this thing online so that I can run team viewer so a hardware vendor can login to the computer and update a software.

I feel like every time I try to connect a linux computer to the internet it’s a huge hassle and takes half a day … today is no different … last time I got this box online I ran the following after messing around with YAST and making sure that the system was setup for DHCP.

Note that for today’s battle the best I was able to do was successfully ping my router, but whenever i tried to ping 8.8.8.8 or other outside ip it would either timeout with nothing or say “host unreachable”

sudo su
netconfig -f update
rm /etc/resolv.conf
rcnetwork restart

This sequence of updating the network removing the resolv.conf file, and restarting the network got me online. But today when I moved my PC to a different building and tried to plug it in things kind of fell apart. I was reading that wicket was interfering with things and to disable it by checking the status of network.service, and wicked.service, then disabling them and starting NetworkManager:

systemctl is-active network.service && systemctl stop network.service
 systemctl is-active wickedd.service && systemctl stop wickedd.service
 systemctl disable wicked.service
 systemctl --force enable NetworkManager.service
 systemctl start network.service

Having done this, Network Manager has not helped making connection and I can’t seem to go back to getting network.service started:


systemctl start network service
>Failed to start network.service: Unit network.service failed to load: No Such file or directory 
rcnetwork restart
>Failed to restart network.target: Operation refused, unit network.target may be request by dependency only

doesn’t seem good to me.

Going into Yast → network devices → Network Settings I get the following message and am unable to edit any of the network devices:

WARNING: Network is currently handled by NetworkManager or completely disabled. YaST is unable to configure some options

also not looking very good …

here’s some quick info for current ip setup:


ip a
1: lo: <loopback, up, lower_up> mtu ... unknown group default
     link/looopback 00:00:00:00:00 brd 00:00:00:00:00:00
        valid_lft forever preferred_lft forever ...

2: eth0: <BROADCAST, MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 100
       link/ether 00:14:d4:.. brd ff:ff:ff:ff:ff:ff
ip r
<no output>

any thoughts on getting back to square one? from there maybe I can start up with trying to reach the outside network.

Also - in case you’re wondering “why am I running a 3 year old SUSE distro?!” the answer is because this computer lives to run a piece of machinery and the tool company installed a custom and expensive software package on this computer. It typically does just fine not being online and running the tool but we just moved the tool and for whatever reason moving the computer to a new building inadvertently messed up everything (possibly a bad HD) but that’s besides the point. Yes we are looking into getting a new computer from the tool manufacturer but for some reason they’re telling us they need to get access (through team viewer) to the original software they made otherwise it will cost 3x as much to get a “from scratch” software made. c’est la vie.

Thanks!

2

Hi
So the cable is ok…?

Just use YaST and wicked and it should be fine with DHCP…

You need to check the ethernet driver is loaded;


/sbin/lspci -nnk | grep -A3 Network

Also check the output from the command dmesg and see it’s not asking for the likes of firmware.

3

“Host unreachable” means that your configured IP address isn’t compatible with the network it’s connected to.
This wouldn’t be surprising if you moved between networks and only suspended the machine during your move, you should shutdown the machine completely and power up again in the new network.

Else, for your machine to pick up new network settings in your new network, you’d have to restart your network service

systemctl restart network.service

When you open YaST > Network Settings and see the warning that your machine is configured to use Network Manager, that’s fine.
You should know that there are two ways to configure network settings… Wicked (traditional ifup/ifdown) configured using YaST or Network Manager.
Whichever way you’re managing your networking, you should know the Global Options tab in your YaST Network Settings module… to the far left. With that, you can switch between Wicked, Network Manager or no networking and should not be done using the command lines you posted. Usually not important, but the Hostnames tab can sometimes be important but not in the situation you’re describing.

Once you’ve set up Network Manager to manage your networking,
You should then look for the Network Manager applet in the “tray” of your Desktop, typically to the far right if your Desktop panel is positioned along the bottom of your screen (most common configuration).
Click on that to display and configure network connections,
You probably have at least one and is configured as a DHCP client.

HTH,
tsu

4

Thanks for the input.

To answer malcolmlewis’ question about the ethernet cable: It should be OK, I literally took the cable out of my other windows desktop (which was getting internet access no problem) and plugged it into the SUSE computer.

When I run


/sbin/lspci -nnk | grep -A3 Network

I don’t get anything: I’m guessing this means that my ethernet driver is not loaded? Running that without the “grep A3 -Network” query shows a list of the devices and controllers, the only one mentioning ethernet is:


02:08.0 Ethernet controller [0200]: Intel Corperation NM10/ICH7 Family LAN controller [8066:27dc] (rev 01)
Subsystem: Hewlett-Packard Company Device [1-3c:2a22]
kernel driver in use: e100
kernel modules: e100

for the dmesg command - there’s a lot of information put out, looking through it the following seem like they might be related but I don’t really know, what should I be looking for?


ACPI Exception: AE_NOT_FOUND, While evaluating sleep state \_S2_] (20140424/hwxface-500)
...

acpi PNP0A0b:00: [Firmware Info]: MMCONFIG for domain 0000 [bus 00-3f] only partially covers this bridge 
...

pc1 0000:02:08.0: Firmware left e100 interrupts enabled; disabling 
...

That’s as much as I could find that seemed related to firmware from the dmesg command.

Regarding YaST and wicket - when I open YaST I get that error I mentioned (This error was new after I ran the *systemctl Force enable NetworkManerger *command) on the global options tab for Network setup method I have the following options in the dropdown:

  • Network Services Disabled
  • Wicked Service
  • NetworkManager Service

I have Enable IPv6 checked

Under overview I have two devices:

Ethernet Network Card | 170.168.10.66| eth0-save
NM10/ICH7 Family LAN Controller | DHCP | eth0

I have no options here to add, edit, or delete the settings of these - it also says “(Not connected)” on the NM10 LAN controller

I feel like I shot myself in the foot when I ran this set of code:

systemctl is-active network.service && systemctl stop network.service
 systemctl is-active wicked.service && systemctl stop wickedd.service
 systemctl disable wicked.service
 systemctl --force enable NetworkManager.service
 systemctl start network.service

before I ran this I still could edit stuff under YaST and had network.service, after I ran that I now have NetworkManager.service and network.service seems to be deleted.

When I run

systemctl restart network.service

I get

Failed to restart network.service: Unit network.service failed to load: No such file or directory

@Tsu2

I did have the machine powered down during the move between buildings but thanks for the info regarding IP address compatibility with my network, I do have a network registration service on my campus which I have to register the device with, maybe that’s one piece of the puzzle but I need to fix network.service first. based on the stuff **malcolmlewis **mentioned I think there’s something up with my firmware and eth0 driver which probably needs to be fixed before moving forward.

5

Hi
So it does need firmware;


/sbin/modinfo e100 | egrep "27DC|firmware"

firmware:       e100/d102e_ucode.bin
firmware:       e100/d101s_ucode.bin
firmware:       e100/d101m_ucode.bin
alias:          pci:v00008086d000027DCsv*sd*bc02sc00i*

Can you check the files exist in the firmware directory;


ls /lib/firmware/e100/
6

Frankly,
I doubt that you have a firmware issue, if your NIC worked in your original network, then your firmware works.
Changing to a new network doesn’t have anything to do with the basic functionality firmware provides, it’s purely a configuration problem.

The above should be replaced with the following

systemctl restart network.service

The above should be replaced by going into YaST > Network Settings,
Then in the left-most “Global Options” tab select your Choice of Wicked, Network Manager or Network Disabled.

Unknown reason for the above error, at least what you posted should not have removed your network.service but if it’s missing I’m pretty sure you won’t have any kind of networking at all… “network.service” is kind of like a master configuration over several other services.
Try the following

systemctl status network.service

and if the above still fails, then run the following command which displays the dependency chain starting with network.service

systemd-analyze critical-chain network.service

It should resemble the following

# systemd-analyze critical-chain network.target
The time after the unit is active or started is printed after the "@" character.
The time the unit takes to start is printed after the "+" character.


network.target @31.481s
└─NetworkManager.service @30.768s +712ms
  └─network-pre.target @30.768s
    └─firewalld.service @17.112s +13.655s
      └─polkit.service @18.674s +5.473s
        └─basic.target @14.884s
          └─paths.target @14.884s
            └─btrfsmaintenance-refresh.path @14.884s
              └─sysinit.target @14.845s
                └─sys-fs-fuse-connections.mount @20.579s +9ms
                  └─systemd-modules-load.service @3.006s +867ms
                    └─systemd-journald.socket
                      └─-.slice

I would also recommend going into the YaST > Network Settings > Global Options tab again as I described above and set Network Manager that way (if necessary, change to something else and then back again to Network Manager) on the chance that your command line attempt to switch from Wicked to NetworkManager might he broken and be reset this way.

The above verifies that your eth networking interface is down which is no surprise.
Hopefully something I suggested above should have fixed your problem, possibly with a reboot.

Otherwise,
At least for the moment I’m almost certain that at least temporarily you can run the following which would bring up your interface… Although it’s a Wicked command, it will work even if you’re configured to use Network Manager.

ifup eth0

HTH,
TSU

7

An additional try which should re-install your networking components and configuration…

First, make sure you have your OSS and Update repos are configured…
I assume that you don’t have either configured for online sources.
If you have any questions about configuring your repos, ASK FIRST… Configuring your repos incorrectly can really mess up your machine.
First disable your existing OSS and OSS-Update repos, they’re probably pointing to repos that don’t exist anymore… You can do this using YaST or zypper.
Next,
Enable the archive online repos for 13.2 by running the following commands (assuming you’re running 64bit), and accept the GPG keys

zypper ar https://ftp5.gwdg.de/pub/opensuse/discontinued/distribution/13.2/repo/oss/suse/ archive_13.2_OSS

zypper ar https://ftp5.gwdg.de/pub/opensuse/discontinued/update/13.2/ archive_13.2_OSS_Update

Followed by a refresh to verify your repos are configured properly and working

zypper ref

Note that the repos are not configured to auto-refresh, there is no point to updating your system regularly since 13.2 is past EOL.

When your repos are working,
Now you can re-install your networking components with the following

zypper in -f yast2-network

If successful, your networking will now be re-installed with the 13.2 default settings, which means it’s set to use Wicked for managing Networking. If you’re going to continue using eth0, then you may want to keep using Wicked, it’s simpler to use than Network Manager, you just go into YaST and instead of the Global Settings tab, edit your Device to have a DHCP configuration… and you’re done.

Else,
If you want to go back to using Network Manager,
Then you’ll want to use Network Manager, then as described above use the Global Settings tab to switch over instead of the command lines you tried.
Then, exit YaST and find the Network Manager applet in your panel tray, click on it and create a Network Connection, make sure it’s configured as a DHCP client and perhaps set it to connect as a system (bootup) service instead of connecting when you logon, and you’ll be working.

HTH,
TSU

8

Here is what I found:


modinfo e100 | egrep "27DC|firmware"

>firmware: e100/d102e_ucode.bin
>firmware: e100/d101s_ucode.bin
>firmware: e100/d101m_ucode.bin
>alias:          pci:v00008086d000027DCsv*sd*bc02sc00i*

However, to your point I don’t think the firmware files are actually there:


ls /lib/firmware/e100
<no result>

Where can I get the firmware files? seems like I need those three d102, d101, etc?

9

Hi
They may be in a different directory in the 13.2 kernel.

Try;


find /lib/firmware/ -name "d10*.bin"
10

You are right - I found the d101m, d101s, and d102e files in:


/lib/firmware/3.16.7-42-default/e100

and

the remaining d10XX_ucode.bin files in:


/lib/firmware/3.16.2-default/e100

So since they’re not missing are they in the wrong location?

11

Working through TSU2’s recommendations here’s where things go:

systemctl restart network.service
Failed to restart network.service: Unit network.service failed to load: No such file or direcotry. 


systemctl status network.service
network.service
Loaded: not-found (Reason: No such file or directory)
Active: inactive (dead)


systemd-analyze critical-chain network.service
The time after the unit is active or started is printed after the "@" character.
The time the unit takes to start is printed after the "+" character.

<no output>

From here, I went into YaST and attempted to change the global option to NetworkManager but it complained saying “Network Manager deamon not running, needs a desktop applet to run, make sure this is on” then proceeded to do nothing.
So I updated the service to wicked and restarted the computer. Now I’m getting the following:

systemctl status network.service
Wicked.service - wicked managed network interfaces
Loaded: loaded (/usr/lib/systemd/system/widked.service; enabled) 
Active: (exited) since Tue ... timestamp
Process: 990  Execstart/usr/sbin/wicked --systemd ifup all (code=exited, status=0/success)
Main FID: 990 (code=exitd, status=0/success)
CGroup: /system.slice/wicked.service

Jul 16 ... : device eth0-save: operation timed out
Jul 16 ... : lo     up
Jul 16 ... : eth0   up
Jul 16 ... : eth0-save  no-device
Jul 16 ... : started wicked managed network interfaces

Still not able to ping 8.8.8.8 - it says “Host unreachable” and I get a timeout on ping google.com.

This is something resembling progress but still not going too far - before I would just get an unknown host error with ping google.com now it just sits there and times out.

12

Quick update as well after switching the service to wicked:

With the eth0 device setup for a statically assigned IP address from my works network registration I’m getting the following status:


systemctl status network.service
Wicked.service - wicked managed network interfaces
Loaded: loaded (/usr/lib/systemd/system/widked.service; enabled) 
Active: (exited) since Tue ... timestamp
Process: 990  Execstart/usr/sbin/wicked --systemd ifup all (code=exited, status=0/success)
Main FID: 990 (code=exitd, status=0/success)
CGroup: /system.slice/wicked.service


Jul 16 ... : eth0   up
Jul 16 ... : eth0-save  no-device
Jul 16 ... : started wicked managed network interfaces
Jul 16 ... : Reloading wicked managed network interfaces
Jul 16 ... : device eth0-save: operation timed out
Jul 16 ... : device eth0-save: operation timed out
Jul 16 ... : Reloading wicked managed network interface
Jul 16 ... : eth0   device-ready
Jul 16 ... : eth0   up
Jul 16 ... : eth0-save  no-device

Trying to ping some things, still getting time out / unknown


ping 8.8.8.8 
... destination host unreachable 
ping google.com
ping: uknown host google.com

I get basically the same result regardless of whether I’m using a DHCP or static IP setup in YaST. through my network adminstrator I have a statially assigned IP address, however, the last time I was able to get this box online I had Yast configured with DCHP and that worked even though we tried with the statically assigned IP address which didn’t reach the internet.

I am getting an output from systemd-analyze though:


# systemd-analyze critical-chain network.service
The time after the unit is active or started is printed after the "@" character.
The time the unit takes to start is printed after the "+" character.


network.target @31.481s
└─wickedd-nanny.service @30.768s +712ms
  └─wickedd.service @30.768s
    └─widkedd-dhcp6.service @17.112s +13.655s
      └─dbus.service.service @18.674s +5.473s
        └─basic.target @14.884s
          └─timers.target @14.884s
            └─systemd-tmpfiles-clean.timer @14.884s
              └─sytinit.target @14.845s
                └─apparmor.service @20.579s +9ms
                  └─systemd-tmpfiles-setup.service @3.006s +867ms
                    └─local-fs.target @16.431s
                       |__ run0user-1000-gyfs.mount @1min 9.9s 
                           |__local-fs-pre.target @003s 
                              |__system-remount-fs.service @3.6s
                                  |_system-readahead-replay.service
                                     |_system.slice
                                        └─-.slice

Any other tips before I go the route TSU2 said about reinstalling all of the network components of the system?

13

Hi
With wicked did you set a static ip address for eth0? If so, you also need to set the default gateway and DNS entries… did you do this?

What is eth0-save? Does it show in the main YaST network setting GUI page?

14

Hi Malcolmlewis - thanks for the reply!

I have tried setting up eth0 as both a static and DHCP IP. Neither have worked for me unfortunately. When I set it statically through YaST I just input the IP address, the subnet mask, and the Hostname - I am not sure how to set the default gateway and DNS entries: pardon me for not knowing, but are these also set in YaST? I don’t see anything talking about DNS or default gateway on the “General”, “Address”, or “Hardware” tabs of the edit Device page in YaST. Under “Hostname/DNS” I see the Hostname and Domain Name have information in them (“linuxDWL” and “site” - values which I haven’t changed since getting into this mess but which seem right?). The Modify DNS configuration is set to “Use Default Policy”.

Is it under the Routing page? I see some IPv4 and IPv6 gateway information and a routing table - everything is empty right now though, is this where I would assign that and where should I find the right gateway string / number?

To answer your question about “what is eth0-save”, that is a LAN connection we make to another computer on the tool which this computer is for, it’s a bit obnoxious but the tool interfaces with a Windows machine, a linux box, and an OS9 computer. the eth0-save is the connection to the OS9 computer.

Thanks

15

Hi
Yes, if setting the static ip address the other information needs to be added, dns servers eg openDNS (208.67.222.222 and 208.67.220.220) and then the ip address of your router as the gateway.

16

Got it: seems like the device assignment was messed up because of the previous networking between the OS9 computer on the tool and the campus network.

Recall I had the eht0-save device setup with a 170.168.11.77 IP address that was statically assigned - this was to connect via ethernet to the OS9 computer. The other device was the NM10 … LAN eth0 which should have been reaching the network and internet.

Seemed like there was a conflict between the two devices and I essentially had to manually remove the eth0-save device configuration with

ifdown eth0

Then assign it to DHCP in YaST then manually bring it back up with

ifup eth0

which finally accepted the network assigned IP address for the machine

At this point the IP address was correct but I still wasn’t able to ping google.com and needed to update the routing by manually adding a default gateway:

route add default gw 128.2.56.1

Probably could have been done through YaST as well but essentially Malcomlewis was right: It was a combination of messed up eth0 device and gateway assignment

Thanks for the replies and help trouble shooting this issue

17

Following your posts,
I can see your progress.

IIRC adding the router by command line as you did won’t be persistent,
]You either need to edit the interface file directly (the old way) or in YaST where you would have used the Global Options tab, you also have a routing tab where you can set your Default Gateway and a Hostnames tab where you can set your DNS servers.

Keep in mind setting up using Wicked and YaST means your network settings will be lost if you take your computer to another network and set up a new network connection unless you can configure as a DHCP client on both networks, then you only have to restart your network service.

If your machine is portable and used in many networks,
You’ll want to set up using Network Manager.

Keep up your good work, you’re making progress and learning…

TSU