Last week (31/5) I posted about my network problems but I didn’t get any answers. I try again:
On my laptop i586/version 20150529/plasma5 after startup I have no internet connection. The problem is no gateway. Internal I can ping, I can open/save files from the NAS within my network but I cannot get out on the internet. So I presume the hardware is OK.
I attempted to set a gateway manually (route command does not work any more so I used the ip route command):
linux-top:~ # ip route
192.168.2.0/24 dev eth0 proto kernel scope link src 192.168.2.12
192.168.2.0/24 dev eth0 proto kernel scope link src 192.168.2.12 metric 100
linux-top:~ # ip route add default via 192.168.2.1
linux-top:~ # ip route
192.168.2.0/24 dev eth0 proto kernel scope link src 192.168.2.12
192.168.2.0/24 dev eth0 proto kernel scope link src 192.168.2.12 metric 100
linux-top:~ #
I tried to set the gateway via Yast and Connection editor - no change.
I tried to get information from wicked
linux-top:~ # rcwicked status
● wicked.service - wicked managed network interfaces
Loaded: loaded (/usr/lib/systemd/system/wicked.service; enabled; vendor preset: disabled)
Active: active (exited) since Sat 2015-06-06 10:46:40 NZST; 6min ago
Process: 818 ExecStart=/usr/sbin/wicked --systemd ifup all (code=exited, status=0/SUCCESS)
Main PID: 818 (code=exited, status=0/SUCCESS)
CGroup: /system.slice/wicked.service
Jun 06 10:46:35 linux-top.site systemd[1]: Starting wicked managed network interfaces...
Jun 06 10:46:40 linux-top.site wicked[818]: **device eth0: call to org.opensuse.Network.Addrconf.ipv4.static.requestLease() failed: General failure**
Jun 06 10:46:40 linux-top.site systemd[1]: Started wicked managed network interfaces.
Jun 06 10:46:40 linux-top.site wicked[818]: lo up
Jun 06 10:46:40 linux-top.site wicked[818]: eth0 setup-in-progress
linux-top:~ # rcwicked restart
linux-top:~ # rcwicked status
● wicked.service - wicked managed network interfaces
Loaded: loaded (/usr/lib/systemd/system/wicked.service; enabled; vendor preset: disabled)
Active: active (exited) since Sat 2015-06-06 10:54:17 NZST; 4s ago
Process: 2619 ExecStop=/usr/sbin/wicked --systemd ifdown all (code=exited, status=0/SUCCESS)
Process: 2622 ExecStart=/usr/sbin/wicked --systemd ifup all (code=exited, status=0/SUCCESS)
Main PID: 2622 (code=exited, status=0/SUCCESS)
Jun 06 10:53:47 linux-top.site systemd[1]: Stopping wicked managed network interfaces...
Jun 06 10:53:47 linux-top.site systemd[1]: Starting wicked managed network interfaces...
Jun 06 10:53:47 linux-top.site wicked[2619]: eth0 device-ready
Jun 06 10:54:17 linux-top.site wicked[2622]: **device eth0: operation timed out**
Jun 06 10:54:17 linux-top.site wicked[2622]: lo up
Jun 06 10:54:17 linux-top.site wicked[2622]: eth0 setup-in-progress
Jun 06 10:54:17 linux-top.site systemd[1]: Started wicked managed network interfaces.
linux-top:~ #
Last time after a hardware check i found that module r8169 is used so I tried modprobe r8169 and afterward - chance or not - the network was up. I tried this and got only an answer - something with FATAL however with lsmod
linux-top:~ # lsmod | grep r8
r8169 86016 0
the module seems to be here. I tried several time rcwicked restart, I rebooted several times, I shut down and left the computer for some time and tried again but today nothing works. Now I went to one of the 13.2 computers which connect without problems to the internet to write this new thread in the hope someone can help me to get the gateway up.
Cheers
No error message, nothing in logs? May be some running networking daemon (wicked?) intercepts it and tries to “fix” unexpected routing entries. Try first to stop any network managing programs and configure IP address and route manually. Does it work?
Thank you for all those who answered. Later yesterday I tried again and this time the connection (including gateway) was up straight away. I don’t know why, in the morning I tried for more than an hour and nothing worked and in the afternoon it was up. I upgraded to version 29150603 and this morning I could log straight in - may be the upgrade solved something. If not I would like to know what the problem is.
arvidjaar unfortunately i don’t know much about the logs. in /var/log/NetworkManager the last line is
2015-05-12T18:24:30.736292+12:00 linux-top nm-dispatcher: Dispatching action 'dhcp4-change'
for eth0
and that looks as if the entry was from 12 May? Date and time are set correctly at the computer and I found nothing (at least where I could make sense of) in boot.log, warn, fail. localmessages. I didn’t try to set a static IP address but I tried to manually (as root) set the gateway which is missing. The rest of the network worked OK.
keellambert I couldn’t find much in YAST - Support - Release on network except some for wireless. The network is managed through wicked
**linux-top:/var/log #** systemctl -p Id show network.service
Id=wicked.service
I didn’t know the command nmtui-connect but now (today when the network is up) it shows only that wireless is not connected and wired is. All I could do there is disconnect and connect again. Is there anything else this command can tell me?
hcvv netstat does not seem to exist any more. I presume that is one of the commands like ifconfig or route which are deprecated?
**linux-top: #** netstat -rn
If 'netstat' is not a typo you can use command-not-found to lookup the package that contain
s it, like this:
cnf netstat
**linux-top: #**
From ‘ip route’ I could see that there is no default gateway, nslookup does not work (even though the correct name servers are in /etc/resolv.conf) and the internal network is working - I have a NAS and a network printer which I can reach and I can ping the other computers.
So has anyone any idea how I can make this connection more reliable that I can use this computer regularly?
Sorry, I am used to it since about 40 years. And as it gives you a much better layout I did hardly try to understand the scribbly text of your route statement. The old rule: something new is not always something better. >:)
Perhaps. It does take getting used to, (especially with respect to the routing table format). FWIW, I had to when I started supporting/configuring (Linux-based) Mikrotik routers which use the ‘ip’ commands. Anyway, I think it’s better get familiar with using iproute2 commands where possible (as with systemd commands). With Linux, one never stops learning.
If two moderators in the forum discuss ifconfig versus ip addr then I am probably out of luck and nobody can suggest how to solve my problem. I will just put together what I think the problem is and if I get no answer we will leave it at that.
Previously I never had a problem connecting to the internet. My router sometimes misses issuing the default gateway but on every computer I could enter the gateway manually with the command ‘route add default gw 192.168.2.1’. Since the router is the requirement from my ISP I cannot change that. The ISP just claims I am the only one having this problem - and if they can have a look at the router but if they find nothing wrong I have to pay for it. So until now - with entering the gateway by command line when no gateway was up - I had no problem. (From around August we will have fibre connection available - that might be the time to change ISP! and that might solve the problem!). I presume the set-up was through network manager.
In May I upgraded to Tumbleweed. Now wicked took over the network configuration. (I take that from the fact that the last line in the NetworkManger log is from the 12 May.) Further the route command was left off the default OS since it is deprecated and the ip route command should be used as default. Suddenly the gateway cannot be put in manually (wicked or ip route command?) any more. rcwicked restart or rebooting did not solve the problem either.
The problem happens at random - I cannot reliably reproduce it. (This seems to be the problem of the router - is the dhcp sequence sometimes timing out before completed?). So at the moment (this morning) I had the internet connection straight away. May be the problem solves itself with an upgrade or by changing ISP with a different router.
Cheers
Uli
]If two moderators in the forum discuss ifconfig versus ip addr then I am probably out of luck and nobody can suggest how to solve my problem.
Well, the ‘net-tools’ and ‘iproute2’ tools both provide the same information. They help to provide a clearer picture to us who are not sitting in front of your computer.
I will just put together what I think the problem is and if I get no answer we will leave it at that.
The more information that you can provide, the better. Verbose descriptions don’t necessarily help with this. Commands and output is what usually counts.
In May I upgraded to Tumbleweed. Now wicked took over the network configuration. (I take that from the fact that the last line in the NetworkManger log is from the 12 May.) Further the route command was left off the default OS since it is deprecated and the ip route command should be used as default. Suddenly the gateway cannot be put in manually (wicked or ip route command?) any more. rcwicked restart or rebooting did not solve the problem either.
You can use ‘ip route’ to add routes and gateways but they won’t be persistent. For example
ip route add default via 192.168.1.1
This will then be shown in the routing table
ip route
Wicked and NM provide the means to configuring persistently, and you need to decide which you want to use.
The problem happens at random - I cannot reliably reproduce it. (This seems to be the problem of the router - is the dhcp sequence sometimes timing out before completed?). So at the moment (this morning) I had the internet connection straight away. May be the problem solves itself with an upgrade or by changing ISP with a different router.
Cheers
Uli
Well, maybe your router is faulty. Difficult to tell from where we sit.
Did you read the original post? The problem is that this apparently has no effect. It may actually be the reason why setting default route from DHCP server fails.
That’s what I wrote and I put the code of several of the (iproute2) commands in the first post of the thread.
You can use ‘ip route’ to add routes and gateways but they won’t be persistent. For example
ip route add default via 192.168.1.1
This will then be shown in the routing table
ip route
The first code in the first post shows how my computer reacted to both - could not set default gateway and consequently the internal network was up but no internet. As written before, I put the entries for the default gateway in YAST and in the Connection editor in an attempt to set it for every log in but this does not help - the default gateway is sometimes not there.
Wicked and NM provide the means to configuring persistently, and you need to decide which you want to use.
Yes and I wrote - that became clear to me in this thread - that with the upgrade to Tumbleweed the machine must have switched from NM to wicked.
Well, maybe your router is faulty. Difficult to tell from where we sit.
Up to now (and on my other machines) I could correct the router’s (dhcp) problem but now not any more.
deano_ferrari - I appreciate you trying to help me and I am sure my descriptions are not the best and for someone trying to imagine what is happening on my computer is not easy. The beginning of my last post was only referring to your and hcvv’s last posts discussing the two types of commands - no offence intended.
Cheers
Uli
I have similar problems on only one of my 3 tumbleweed machines too - I can use NM or wicked that makes no difference. Randomly the network is up after boot or not. I can manualy start e.g. “sudo NetworkManager” and it is then up. And yes, I have only wicked or only NM enabled at the same time through the yast configuration.
If it is not up after boot, then the boot log says things like:
systemd[1]: Job
NetworkManager-wait-online.service/start deleted to break ordering
cycle starting with network.target/start
I think this all started way back when systemd was upgradet to version 219. I did not report this earlier because first I thought my hardware was at fault (motherboard is 7 years old) but on windows the problem never shows and after I found out that sudo NetworkManager is a quick fix, I just hoped that with some update the problem will go away eventually…
I think your issue is apparently related to systemd issues. I have no idea why the repetitive change of state is reported with NetworkManager. Probably best dealt with via a bug report.
I’m not good with this sort of stuff, but I do wonder why your system thinks the interface is eth0 when it should perhaps be using a systemd kind of name like mine: enp0s25
