Adding a permanent route

jbailo · October 30, 2008, 11:04pm

**Hm, and no errors during boot (use dmesg)? **

I see this error:

ADDRCONF(NETDEV_UP): eth0: link is not ready
tg3: eth0: Link is up at 100 Mbps, full duplex.
tg3: eth0: Flow control is off for TX and off for RX.
ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
eth0: no IPv6 routers present
bootsplash: status on console 0 changed to on
ip6_tables: (C) 2000-2006 Netfilter Core Team
ip_tables: (C) 2000-2006 Netfilter Core Team
nscd[2396]: segfault at f80012f4 ip b7ff56f4 sp afc39fd8 error 4 in nscd[b7fe5000+1b000]

malcolmlewis · October 30, 2008, 11:35pm

Hi
What windows OS? I don’t use windows on a regular basis… but do have
XP.

–
Cheers Malcolm °¿° (Linux Counter #276890)
openSUSE 11.0 x86 Kernel 2.6.25.18-0.2-default
up 21:48, 1 user, load average: 0.31, 0.40, 0.29
GPU GeForce 6600 TE/6200 TE - Driver Version: 177.80

hcvv · October 31, 2008, 9:14am

I should have tried this earlier. I added a route with YaST on my system and then looked what happend in /etc/sysconfig/network/routes.

It made

10.40.1.0 192.168.26.254 255.255.255.0 eth0

and it put it **before **the default.

BTW, I could not remove it with YaST after the test. It was there, selecting it did not activate the Delete nor the Edit button. So I removed the line from /etc/sysconfig/network/routes and hopes that my system is not screwed up

jbailo · October 31, 2008, 7:41pm

What windows OS?

Using XP Pro for NX client. The readme says that Ctrl-Alt-F will not work in XP for the 2.x client, but I’m using the 3.x client. I checked in the About that it’s 3.x

jbailo · October 31, 2008, 7:44pm

I find that if I use the route command I have to specify the device (**eth1 **in my case, I don’t use the eth0 card).

However, I did not put a device in my *routes *file because I was following the pattern in there already and what I read here. Perhaps I need to put eth1 in in my *routes *file entry.

malcolmlewis · November 2, 2008, 12:33am

Hi
Just tried the 3.x client in XP and you are correct about not being
able to switch to full screen. Might I suggest just starting the client
in full screen mode, then minimizing?

–
Cheers Malcolm °¿° (Linux Counter #276890)
openSUSE 11.0 x86 Kernel 2.6.25.18-0.2-default
up 0:08, 1 user, load average: 0.30, 0.62, 0.46
GPU GeForce 6600 TE/6200 TE - Driver Version: 177.80

jbailo · November 4, 2008, 7:17pm

I hate to tell you this…but I found an answer.

My Yast screen does have those Global Options.

The problem was my terminal window was sized in such a way that the top menus were cut off! I don’t know what I was thinking but somehow I thought it would size itself to show everything.

When I maximized my (putty) terminal window, suddenly I could see the addition options you’ve been telling me about and could find how to set routes!

Thanks again, and sorry to waste your time (but also the information about NX is really valuable).

hcvv · November 4, 2008, 7:38pm

Thanks for telling us. I could not understand what was wrong. and it is always nice to hear tthe solution.

All the best.

jbailo · November 10, 2008, 9:34pm

I finally got around to try and add these routes now that I can see the menus in Yast.

But now I have a new problem!

While I can enter a Default Gateway, and I can see the screen area for Routing Table, the menu items for Add|Edit|Delete are not active!

I cannot tab to the Add command and so I cannot add any routes! And there is no yellow “A” highlighted, so I can’t use the keyboard command either!

Is there some permissions problem?

I logged in with my user account and then did “su -” to bring up yast.

hcvv · November 10, 2008, 10:23pm

Are you sure you have ‘Traditional Method with ifup’ in the Global Tab?

When I change this to ‘User controled …Networkmanager’ The Add button greys out.

jbailo · November 11, 2008, 12:15am

Ok, but will that change all my network settings?

I use a static IP and so on…and the server is not somewhere where I can get access to the console if I screw it up.

hcvv · November 11, 2008, 9:52am

Well, I should think this will work OK, but I can not garantee

At least after changing the bullet to ifup, check all the tabs if everything is there and OK before acceptence.

I also think that proof of not introducing any problems is only given after first reboot. I admit that this is all a bit tedeous with remote system management.

I only use ifup, even on my desktops, and IMHO it is the better method to use in server like systems. I never understood what a network manager has to do on such a system. Once the system is running everything is static. Of course things are different on laptop like systems with wifi connections that are moved around and change their configuration time and again.

jbailo · November 11, 2008, 7:43pm

Well seriously, I am not sure even why I need these “routes”. My ISP put them in, and I guess I understand that they allow for more efficient travel to commonly used subnets and nodes, but this server is on a small subnet behind a firewall. It’s not going to make that many hops any way…I think…

hcvv · November 11, 2008, 8:36pm

Well, as long as you can reach all systems I see no real need to add those routes. And when nobody complains about performance …

Normally the efficient routing through the net is the task of the routers, not of the end-systems.

That said I should think about converting them to traditional ifup, if you ever have a maintenance window at the system, if I where you. It is quickly done and tested. And easy to revert if you are in touch with the metal.

I also would see that a new system installation would use that from the start.

jbailo · November 11, 2008, 11:22pm

Well, changing it from Network Manager to Traditional Method in Yast disconnected it from the network.

Now I’m having someone restore it remotely.

Sigh…

hcvv · November 12, 2008, 8:40am

Oops I jopes you would do it when being at the system because no need to do it now.

All the best.

jbailo · November 12, 2008, 11:40pm

Here’s what puzzles me.

I am looking at what I think needs to be in place for me to switch from Network Manager to ifup.

I have two network cards, the one I’m using for network connectivity is eth1. It’s config is:

# cat ifcfg-eth1
BOOTPROTO='static'
BROADCAST=''
ETHTOOL_OPTIONS=''
IPADDR='192.168.26.201/24'
MTU=''
NAME='NetXtreme BCM5701 Gigabit Ethernet'
NETWORK=''
REMOTE_IPADDR=''
STARTMODE='auto'
USERCONTROL='no'

What puzzles me here is why Network Manager configures IP as

IPADDR='192.168.26.201/24'

and not

IPADDR='192.168.26.201'

Moving on then my DNS is correct:

# cat /etc/resolv.conf
nameserver 216.183.108.115
nameserver 63.91.110.178
search site

And my gateway is set:

# cat routes
default 192.168.26.1 - -

So, I would think that switching to “ifup” in Yast would be all that was needed.

Or am I supposed to an

ifup eth0

After making the change in YaST?

ken_yap · November 13, 2008, 12:07am

Don’t know about the other stuff, but the IPADDR setting

IPADDR=‘192.168.26.201/24’

is just shorthand for two settings at once, address=192.168.26.201 and netmask=255.255.255.0

jbailo · November 13, 2008, 6:43pm

Ok, thanks.

Then, as far as I can see, my settings are 100% correct.

However, when I change the selection from Network Manager to Traditional Method using IFUP my system goes offline.

That’s why I ask, am I supposed to change anything else or issue some other command when changing that option…or do I have to reboot, etc…

jbailo · November 13, 2008, 7:35pm

Here’s something else I don’t understand. My server has two NICs. A built in one (eth0), and a high capacity Netextreme card (eth1).

So, I set up the high capacity card with the static address I want. I didn’t do any manual configuration to the built-in one.

But when I run ifconfig -a I see:


ivnwmail:/etc/sysconfig/network # ifconfig -a
eth0      Link encap:Ethernet  HWaddr 00:06:5B:F7:23:A8
          inet addr:192.168.26.201  Bcast:192.168.26.255  Mask:255.255.255.0
          inet6 addr: fe80::206:5bff:fef7:23a8/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:78034 errors:0 dropped:0 overruns:0 frame:0
          TX packets:90655 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:7702578 (7.3 Mb)  TX bytes:7955882 (7.5 Mb)
          Interrupt:28

eth1      Link encap:Ethernet  HWaddr 00:06:5B:F7:23:A9
          UP BROADCAST MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)
          Interrupt:29

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:42259 errors:0 dropped:0 overruns:0 frame:0
          TX packets:42259 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:39428567 (37.6 Mb)  TX bytes:39428567 (37.6 Mb)

Am I reading this wrong? It looks like eth0 got the IP address and eth1 hasn’t been assigned an address and doesn’t seem to be carrying any traffic either!