Hello all
I am setting up a Centos 6 machine with one network card & one onboard network port. Both are recognized & work. Onboard Network port is unused yet, but is required for dedicated access to LTSP LAN, which would allow older PIII machines to boot from this system. I have installed LTSP but again Networking seems to be the problem area, as the clients won't boot from this machine.
Have muddled & tinkered with Networking a bit...deleted & reconfigured the ifcfg files again & again but as I don't understand NetworkManager and Google is not helping get relevant results, so I have brought the system to minimum working condition, where one network card is providing LAN access on which Internet Router is configured. And now I am posting my system's condition below as output of various commands.
My network-scripts directory:
[root@swyam network-scripts]# ls ifcfg-Internet_GW ifup-eth ifcfg-lo ifup-ippp ifdown ifup-ipv6 ifdown-bnep ifup-isdn ifdown-eth ifup-plip ifdown-ippp ifup-plusb ifdown-ipv6 ifup-post ifdown-isdn ifup-ppp ifdown-post ifup-routes ifdown-ppp ifup-sit ifdown-routes ifup-tunnel ifdown-sit ifup-wireless ifdown-tunnel init.ipv6-global ifup net.hotplug ifup-aliases network-functions ifup-bnep network-functions-ipv6
Ifconfig output:
[root@swyam network-scripts]# ifconfig eth0 Link encap:Ethernet HWaddr E0:69:95:77:A7:F3 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:20 Memory:fe600000-fe620000
eth1 Link encap:Ethernet HWaddr 00:60:52:0C:0B:DE inet addr:192.168.1.3 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2245698 errors:0 dropped:0 overruns:0 frame:0 TX packets:1785431 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:3105161344 (2.8 GiB) TX bytes:148843277 (141.9 MiB) Interrupt:16 Base address:0x2000
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:525 errors:0 dropped:0 overruns:0 frame:0 TX packets:525 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:53459 (52.2 KiB) TX bytes:53459 (52.2 KiB)
[root@swyam network-scripts]# cat ifcfg-Internet_GW TYPE=Ethernet BOOTPROTO=none IPADDR=192.168.1.3 PREFIX=24 GATEWAY=192.168.1.1 DNS1=218.248.245.1 DNS2=8.8.8.8 DOMAIN="218.248.245.1 8.8.8.8 8.8.4.4" DEFROUTE=yes IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME="Internet GW" UUID=d4535709-a05b-459d-a4fd-f08ec9460bcc ONBOOT=yes DNS3=8.8.4.4
Network Manager is running and the above card "Internet_GW" was configured through the NetworkManager applet. Internet works. One can disconnect Internet through the NetworkManager applet. Yet the applet's Edit Connection --> Wired --> Last Used status shows "Never".
Now let's come to what I want to do with this machine....The Onboard network port for LTSP services....Network card on network 192.168.1.0/24(192.168.1.1 is the ISP ADSL router) to provide LAN access & Internet access.
Now what I want to do is to create a third network for my virtual machines (KVM) say 172.16.1.1/24. Now create some sort of dummy bridge and get these machines talking to the Internet....said bridge to either forward the packets through the normal Internet path or somehow use NAT to do it. I think this should be doable but haven't come across any example or any tutorial for this.
Now comes why I want a separate network....why not use 192.168.1.0/24...problem is...I don't want to disable Network Manager by putting Managed=NO in my ifcfg file. As with Network Manager enabled, I can use USB devices whenever my ISP goes down....and I'm afraid it happens more frequently than I care to bear.
I frankly also don't understand how Network Manager works....in a broad manner....literature I found was either too technical or too old or was focussed on a particular problem.
I would like to know a network/bridge/device topology that can solve my issue....three networks...lan. ltsp, virtual machines, all having access to each other & the net, while keeping the option of using networkmanager or any other tool (I'm open) to plugin any other wireless internet access device.
I am not very advanced admin, but can follow instructions.
Kindly help.
With regards & thanks. Sanjay.
On 05/17/2012 10:51 PM, Sanjay Arora wrote:
Hello all
I am setting up a Centos 6 machine with one network card& one onboard network port. Both are recognized& work. Onboard Network port is unused yet, but is required for dedicated access to LTSP LAN, which would allow older PIII machines to boot from this system. I have installed LTSP but again Networking seems to be the problem area, as the clients won't boot from this machine.
Have muddled& tinkered with Networking a bit...deleted& reconfigured the ifcfg files again& again but as I don't understand NetworkManager and Google is not helping get relevant results, so I have brought the system to minimum working condition, where one network card is providing LAN access on which Internet Router is configured. And now I am posting my system's condition below as output of various commands.
My network-scripts directory:
[root@swyam network-scripts]# ls ifcfg-Internet_GW ifup-eth ifcfg-lo ifup-ippp ifdown ifup-ipv6 ifdown-bnep ifup-isdn ifdown-eth ifup-plip ifdown-ippp ifup-plusb ifdown-ipv6 ifup-post ifdown-isdn ifup-ppp ifdown-post ifup-routes ifdown-ppp ifup-sit ifdown-routes ifup-tunnel ifdown-sit ifup-wireless ifdown-tunnel init.ipv6-global ifup net.hotplug ifup-aliases network-functions ifup-bnep network-functions-ipv6
Ifconfig output:
[root@swyam network-scripts]# ifconfig eth0 Link encap:Ethernet HWaddr E0:69:95:77:A7:F3 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:20 Memory:fe600000-fe620000
eth1 Link encap:Ethernet HWaddr 00:60:52:0C:0B:DE inet addr:192.168.1.3 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2245698 errors:0 dropped:0 overruns:0 frame:0 TX packets:1785431 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:3105161344 (2.8 GiB) TX bytes:148843277 (141.9 MiB) Interrupt:16 Base address:0x2000
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:525 errors:0 dropped:0 overruns:0 frame:0 TX packets:525 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:53459 (52.2 KiB) TX bytes:53459 (52.2 KiB)
[root@swyam network-scripts]# cat ifcfg-Internet_GW TYPE=Ethernet BOOTPROTO=none IPADDR=192.168.1.3 PREFIX=24 GATEWAY=192.168.1.1 DNS1=218.248.245.1 DNS2=8.8.8.8 DOMAIN="218.248.245.1 8.8.8.8 8.8.4.4" DEFROUTE=yes IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME="Internet GW" UUID=d4535709-a05b-459d-a4fd-f08ec9460bcc ONBOOT=yes DNS3=8.8.4.4
Network Manager is running and the above card "Internet_GW" was configured through the NetworkManager applet. Internet works. One can disconnect Internet through the NetworkManager applet. Yet the applet's Edit Connection --> Wired --> Last Used status shows "Never".
Now let's come to what I want to do with this machine....The Onboard network port for LTSP services....Network card on network 192.168.1.0/24(192.168.1.1 is the ISP ADSL router) to provide LAN access& Internet access.
Now what I want to do is to create a third network for my virtual machines (KVM) say 172.16.1.1/24. Now create some sort of dummy bridge and get these machines talking to the Internet....said bridge to either forward the packets through the normal Internet path or somehow use NAT to do it. I think this should be doable but haven't come across any example or any tutorial for this.
Now comes why I want a separate network....why not use 192.168.1.0/24...problem is...I don't want to disable Network Manager by putting Managed=NO in my ifcfg file. As with Network Manager enabled, I can use USB devices whenever my ISP goes down....and I'm afraid it happens more frequently than I care to bear.
I frankly also don't understand how Network Manager works....in a broad manner....literature I found was either too technical or too old or was focussed on a particular problem.
I would like to know a network/bridge/device topology that can solve my issue....three networks...lan. ltsp, virtual machines, all having access to each other& the net, while keeping the option of using networkmanager or any other tool (I'm open) to plugin any other wireless internet access device.
I am not very advanced admin, but can follow instructions.
Kindly help.
NetworkManager does not (as far as I know) use ifcfg files unless you select "Available for all users" checkbox, but rather keeps interface data somewhere inside $HOME directory (.gconf subdirectory maybe) in unreadable format for humans.
"Available for all users" will help you, or run "network" service and disable "NetworkManager" service.
Ljubomir Ljubojevic wrote:
NetworkManager does not (as far as I know) use ifcfg files unless you select "Available for all users" checkbox, but rather keeps interface data somewhere inside $HOME directory (.gconf subdirectory maybe) in unreadable format for humans.
Is this true? I have the impression that NM takes note of NM_MANAGED=yes or no in /etc/sysconfig/network-scripts/ifcfg-wlan0 (or whatever). So I assume from this that NM must read these files.
I wish someone would document NM, explaining what files it consults, and something about how it is supposed to function.
In my experience NM works fine 95% of the time; but when it does not work it is often very difficult to work out exactly where the problem lies. In particular the entries in /var/log/messages are completely useless.
Another annoyance is that the network service seems to have got tangled up with NM, at least in Fedora-16; they used to be more or less independent.
On 05/21/2012 03:26 PM, Timothy Murphy wrote:
Ljubomir Ljubojevic wrote:
NetworkManager does not (as far as I know) use ifcfg files unless you select "Available for all users" checkbox, but rather keeps interface data somewhere inside $HOME directory (.gconf subdirectory maybe) in unreadable format for humans.
Is this true? I have the impression that NM takes note of NM_MANAGED=yes or no in /etc/sysconfig/network-scripts/ifcfg-wlan0 (or whatever). So I assume from this that NM must read these files.
Somewhere on the net, undefined time ago (last 2-3 months?) I was asked same question and I investigated on my laptop that has NetworkManager service enabled. But I can not remember when/where I posted results. I think it was in "Linux" Facebook group but search gives nothing.
Do not take my word for it, create NM connection and observe what is happening. And I will run my laptop tomorrow and see if I wrote the search results somewhere.
-
Ljubomir Ljubojevic -
Sanjay Arora -
Timothy Murphy