Bugzilla – Bug 141422
second network card up but not running
Last modified: 2006-02-06 02:43:57 UTC
goal: use one NIC in external zone (internet) and use second card in internal zone (samba, ftp, CUPS). BOTH CARDS WILL USE ONE GATEWAY 192.168.1.1 OS version: SuSE linux 10.0 Hardware: AMD64 3000+ RAM: 2GB HDD: SATA (Seagate) video: nVidia GeForce 6600GT NIC: onboard eth0 - CK8S eth2 - Marvell 88E8001 PCI Linksys NC100 Problem: only eth0 can connect with network problem routing table not updated working setup = eth0 only eth1 only or eth2 only does not work., These cards do not work with combination with eth0 and in fact activation of eth1 or eth2 cloces connection for eth0. exactly the same hardware and network setup worked under SuSE 9.2 when second card is added SuSE looses network connection completely: # ifconfig eth0 Link encap:Ethernet HWaddr 00:0F:EA:7D:94:80 inet addr:192.168.1.8 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:68963 errors:0 dropped:0 overruns:0 frame:0 TX packets:56468 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:80993289 (77.2 Mb) TX bytes:8606806 (8.2 Mb) Interrupt:193 Base address:0xa000 eth2 Link encap:Ethernet HWaddr 00:0F:EA:7D:94:81 inet addr:192.168.1.7 Bcast:192.168.1.255 Mask:255.255.255.0 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:217 Memory:fc020000-0 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:2318 errors:0 dropped:0 overruns:0 frame:0 TX packets:2318 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:335875 (328.0 Kb) TX bytes:335875 (328.0 Kb) 1) eth0 is shown as RUNNING eth2 is not. Also eht1 is NOT RUNNING 2) routing table problem: a) only eth0 # route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 eth0 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo 0.0.0.0 192.168.1.1 0.0.0.0 UG 0 0 0 eth0 b) eth0 + eth2 route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth2 169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 eth0 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo 0.0.0.0 192.168.1.1 0.0.0.0 UG 0 0 0 eth2 no gateway listed for eth0 command: #route add default gw 192.168.1.1 dev eth0 restores routing and network to eth0 however # ifstatus eth2 eth2 device: Marvell Technology Group Ltd. 88E8001 Gigabit Ethernet Controller (rev 13) eth2 configuration: eth-id-00:0f:ea:7d:94:81 eth2 is up 5: eth2: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:0f:ea:7d:94:81 brd ff:ff:ff:ff:ff:ff inet 192.168.1.7/24 brd 192.168.1.255 scope global eth2 eth2 IP address: 192.168.1.7/24 Configured routes for interface eth2: default 192.168.1.1 - - Active routes for interface eth2: 192.168.1.0/24 proto kernel scope link src 192.168.1.7 default via 192.168.1.1 0 of 1 configured routes for interface eth2 up ------------------------------------------------------------------------------------------ # ifstatus eth0 eth0 device: nVidia Corporation CK8S Ethernet Controller (rev a2) eth0 configuration: eth-id-00:0f:ea:7d:94:80 eth0 is up 3: eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:0f:ea:7d:94:80 brd ff:ff:ff:ff:ff:ff inet 192.168.1.8/24 brd 192.168.1.255 scope global eth0 eth0 IP address: 192.168.1.8/24 Configured routes for interface eth0: default 192.168.1.1 - - 169.254.0.0 - 255.255.0.0 eth0 Active routes for interface eth0: 192.168.1.0/24 proto kernel scope link src 192.168.1.8 169.254.0.0/16 scope link default via 192.168.1.1 1 of 2 configured routes for interface eth0 up shows that routing is not configured for eth2 I am unable to ping eth2 from another machine. When only eth2 is up the above problem persists. All three network cards worked in the above setup. Linksys (eth1) has exactly the same problem as Marvell. I did test linksys on another computer and it worked. lsmod shows that: sk98lin, forcedeth and tulip modules are up. ---------------------------------------------------- # getcfg eth0 HWD_HWDESC=eth0; HWD_HWDESCTYPE=interface; HWD_INTERFACE_N=1; HWD_INTERFACE_0=eth0; HWD_INTERFACEPATH_0=/sys/class/net/eth0; HWD_INTERFACETYPE_0=eth; HWD_ID=00:0f:ea:7d:94:80; HWD_BUS_N=1; HWD_BUSNAME_0=pci; HWD_BUSID_0=0000:00:05.0; HWD_VENDORID=0x10de; HWD_PRODUCTID=0x00df; HWD_VENDORSUBID=0x1458; HWD_PRODUCTSUBID=0xe000; HWD_DRIVER=forcedeth; HWD_CONFIG_N=1; HWD_CONFIG_0=eth-id-00:0f:ea:7d:94:80; HWD_DEVICEPATH=/sys/devices/pci0000:00/0000:00:05.0; export HWD_HWDESC HWD_HWDESCTYPE HWD_INTERFACE_N HWD_INTERFACE_0 HWD_INTERFACEPATH_0 HWD_INTERFACETYPE_0 HWD_ID HWD_BUS_N HWD_BUSNAME_0 HWD_BUSID_0 HWD_VENDORID HWD_PRODUCTID HWD_VENDORSUBID HWD_PRODUCTSUBID HWD_DRIVER HWD_CONFIG_N HWD_CONFIG_0 HWD_DEVICEPATH; # getcfg eth2 HWD_HWDESC=eth2; HWD_HWDESCTYPE=interface; HWD_INTERFACE_N=1; HWD_INTERFACE_0=eth2; HWD_INTERFACEPATH_0=/sys/class/net/eth2; HWD_INTERFACETYPE_0=eth; HWD_ID=00:0f:ea:7d:94:81; HWD_BUS_N=1; HWD_BUSNAME_0=pci; HWD_BUSID_0=0000:02:0b.0; HWD_VENDORID=0x11ab; HWD_PRODUCTID=0x4320; HWD_VENDORSUBID=0x1458; HWD_PRODUCTSUBID=0xe000; HWD_DRIVER=sk98lin; HWD_CONFIG_N=1; HWD_CONFIG_0=eth-id-00:0f:ea:7d:94:81; HWD_DEVICEPATH=/sys/devices/pci0000:00/0000:00:0e.0/0000:02:0b.0; export HWD_HWDESC HWD_HWDESCTYPE HWD_INTERFACE_N HWD_INTERFACE_0 HWD_INTERFACEPATH_0 HWD_INTERFACETYPE_0 HWD_ID HWD_BUS_N HWD_BUSNAME_0 HWD_BUSID_0 HWD_VENDORID HWD_PRODUCTID HWD_VENDORSUBID HWD_PRODUCTSUBID HWD_DRIVER HWD_CONFIG_N HWD_CONFIG_0 HWD_DEVICEPATH; -------------------------------------------------- # hwinfo --netcard 16: PCI 05.0: 0200 Ethernet controller [Created at pci.277] UDI: /org/freedesktop/Hal/devices/pci_10de_df Unique ID: rBUF.10LYQ5_nwn8 SysFS ID: /devices/pci0000:00/0000:00:05.0 SysFS BusID: 0000:00:05.0 Hardware Class: network Model: "Giga-byte CK8S Ethernet Controller" Vendor: pci 0x10de "nVidia Corporation" Device: pci 0x00df "CK8S Ethernet Controller" SubVendor: pci 0x1458 "Giga-byte Technology" SubDevice: pci 0xe000 Revision: 0xa2 Driver: "forcedeth" Device File: eth0 Memory Range: 0xfd000000-0xfd000fff (rw,non-prefetchable) I/O Ports: 0xb800-0xb807 (rw) IRQ: 177 (4103076 events) HW Address: 00:0f:ea:7d:94:80 Link detected: no Module Alias: "pci:v000010DEd000000DFsv00001458sd0000E000bc06sc80i00" Driver Info #0: Driver Status: forcedeth is active Driver Activation Cmd: "modprobe forcedeth" Config Status: cfg=no, avail=yes, need=no, active=unknown 27: PCI 20a.0: 0200 Ethernet controller [Created at pci.277] UDI: /org/freedesktop/Hal/devices/pci_1317_985 Unique ID: JNkJ.IH1TcNNtXe0 Parent ID: vuMS.fnpiWG08KJ2 SysFS ID: /devices/pci0000:00/0000:00:0e.0/0000:02:0a.0 SysFS BusID: 0000:02:0a.0 Hardware Class: network Model: "Linksys NC100 Network Everywhere Fast Ethernet 10/100" Vendor: pci 0x1317 "Linksys" Device: pci 0x0985 "NC100 Network Everywhere Fast Ethernet 10/100" SubVendor: pci 0x1317 "Linksys" SubDevice: pci 0x0574 Revision: 0x11 Driver: "tulip" Device File: eth1 I/O Ports: 0x9000-0xafff (rw) Memory Range: 0xfc000000-0xfc0003ff (rw,non-prefetchable) Memory Range: 0x88080000-0x880bffff (ro,prefetchable,disabled) IRQ: 209 (no events) HW Address: 00:4c:69:6e:75:79 Module Alias: "pci:v00001317d00000985sv00001317sd00000574bc02sc00i00" Driver Info #0: Driver Status: tulip is active Driver Activation Cmd: "modprobe tulip" Config Status: cfg=no, avail=yes, need=no, active=unknown Attached to: #21 (PCI bridge) 28: PCI 20b.0: 0200 Ethernet controller [Created at pci.277] UDI: /org/freedesktop/Hal/devices/pci_11ab_4320 Unique ID: mY_N.4aJKZ3Qxce3 Parent ID: vuMS.fnpiWG08KJ2 SysFS ID: /devices/pci0000:00/0000:00:0e.0/0000:02:0b.0 SysFS BusID: 0000:02:0b.0 Hardware Class: network Model: "Giga-byte Marvell 88E8001 Gigabit Ethernet Controller (Gigabyte)" Vendor: pci 0x11ab "Marvell Technology Group Ltd." Device: pci 0x4320 "88E8001 Gigabit Ethernet Controller" SubVendor: pci 0x1458 "Giga-byte Technology" SubDevice: pci 0xe000 "Marvell 88E8001 Gigabit Ethernet Controller (Gigabyte)" Revision: 0x13 Driver: "sk98lin" Device File: eth2 Memory Range: 0xfc020000-0xfc023fff (rw,non-prefetchable) I/O Ports: 0x9400-0x94ff (rw) Memory Range: 0x880c0000-0x880dffff (ro,prefetchable,disabled) IRQ: 217 (23 events) HW Address: 00:0f:ea:7d:94:81 Link detected: no Module Alias: "pci:v000011ABd00004320sv00001458sd0000E000bc02sc00i00" Driver Info #0: Driver Status: sk98lin is active Driver Activation Cmd: "modprobe sk98lin" Config Status: cfg=no, avail=yes, need=no, active=unknown Attached to: #21 (PCI bridge)
Please attach your configuration files in /etc/sysconfig/network. ifcfg-*, ifroute-* and routes. Do i understand that correctly that you want to have all three interfaces in one physical subnet?
O.K. I will be back at home tomorrow evening, I will attach required filkes. "Do i understand that correctly that you want to have all three interfaces in one physical subnet?" Yes, this would be great (worked in 9.2), however, I can't even change NIC to Marvell or Linksys (single card, other NIC disabled). Problem is still there. Currently I can configure only CK8S. Single Marvell or single Linksys do not work.
Your descriptions are confusing. First you had problems with multiple cards, now you say: "Single Marvell or single Linksys do not work." If you have problems in using just one of these NICs with the other two deactivated, then we have to debug this first. I need to know if the interface is not set up correctly (wrong address, routes, etc.) or if it is posibly a driver bug (interface set up correctly but does not work). So please divide your problem into manageable parts.
"I need to know if the interface is not set up correctly (wrong address, routes, etc.) or if it is posibly a driver bug (interface set up correctly but does not work)" Please see above ifstatus and ifconfig. In fact all the information you will find above (default route NIC config). I believe that this problem has nothing to do with bad configuration. If that would be the case How one card could affect another? In other words how incorrectly configured eth2 could cause to stop working eth10. I also explained routing problem. Thank you for your assistance, if you need more information, let me know. Anyway: /etc/sysconfig/network/routes efault 192.168.1.1 - - /etc/sysconfig/network/ifroute-lo 127/8 /etc/sysconfig/network/ifcfg-eth-id-00:0f:ea:7d:94:80 BOOTPROTO='static' BROADCAST='' IPADDR='192.168.1.3' MTU='1500' NAME='Giga-byte CK8S Ethernet Controller' NETMASK='255.255.255.0' NETWORK='' REMOTE_IPADDR='' STARTMODE='auto' UNIQUE='rBUF.10LYQ5_nwn8' USERCONTROL='no' _nm_name='bus-pci-0000:00:05.0' /etc/sysconfig/network/ifcfg-eth-id-00:0f:ea:7d:94:81 BOOTPROTO='static' BROADCAST='' IPADDR='192.168.1.7' MTU='1500' NAME='Giga-byte Marvell 88E8001 Gigabit Ethernet Controller (Gigabyte)' NETMASK='255.255.255.0' NETWORK='' REMOTE_IPADDR='' STARTMODE='auto' UNIQUE='mY_N.4aJKZ3Qxce3' USERCONTROL='no' _nm_name='bus-pci-0000:02:0b.0'
This bugreport is invalid since it mixes different possible problems. I'm not able to see something that goes wrong. Test your devices one by one, if one does not work, file a bug report which deals only with that. Then test your devices all together and if it does not work describe the problem in an easy reproducible way. We have a lot of other work and cannot spend to much time in guessing what the reporter tries to say. If you set up multiple interfaces in the same subnet you have to specify which interface should get the default route. If you don't specify that, then always the interface that has been set up latest will own the default route. Either specify the interface name in 'routes' or use ifroute-* (See the manpages).
no, there is only one problem here. It really does not matter one NIC or more. I did pointed out that for some reason Giga-byte CK8S if single will work Single Marvell or single Linksys will not. I ecxplained where is the error. I also explained that if Giga-byte CK8S will run with second card, problem aggravates and Giga-byte CK8S will lose connection (same reason routing table problem). First you have not read my original post carefully enough and you did ask me to repeat exactly the same information twice, Now you are claiming that there are two different bugs so this makes whole story invalid. Unfortunately your explanation is useless (about routes), I have shown that default route is always the same. If you can't help, either say so or leave this open. It would be more honest. I can't do anything about leaving this or removing or making "invalid" however this will not change anything if problem persist. By the way check SuSe forums to see that I am not only one with similar problem.