Matt Fisher wrote: > > I'm just catching up on this thread. I have two 3c905s in one of my Linux > boxes. I have the 3com support compiled into the kernel (not loaded as a > module), and had to add 'append="ether=0,0,eth1"' (with the single quotes) to > my lilo.conf get to it to recognize both nics. I think there is a similar > string that is added someplace else if you are using modules.
Okay, I tried your method, and right or wrong, it doesn't change my symptoms, which I realized once again I did not elaborate fully. The modules come up on the server so that each of the 3c900b NICs, fully configured, are seen internally and can be pinged from the machine they are on. But, when the two are present, I cannot ping those IP addresses from the networks they connect to. When just one is present, I can get in to the machine from the net, and everything works. The dmesg lines are basically as follows: Adding Swap: 173872k swap-space (priority -1) ne.c:v1.10 9/23/94 Donald Becker ([EMAIL PROTECTED]) NE*000 ethercard probe at 0x240: 00 50 ba a0 92 bb eth0: NE2000 found at 0x240, using IRQ 5. eth1: 3c509 at 0x320 tag 1, BNC port, address 00 a0 24 12 64 4b, IRQ 7. 3c509.c:1.16 (2.2) 2/3/98 [EMAIL PROTECTED] 3c59x.c:v0.99H 12Jun00 Donald Becker and others http://www.scyld.com/network/vortex.html eth2: 3Com 3c900 Cyclone 10Mbps TPO at 0xe400, 00:01:02:6b:43:33, IRQ 9 8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface. MII transceiver found at address 24, status 1809. Enabling bus-master transmits and whole-frame receives. eth3: 3Com 3c900 Cyclone 10Mbps TPO at 0xe800, 00:50:da:b4:4b:5a, IRQ 10 8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface. MII transceiver found at address 24, status 1809. Enabling bus-master transmits and whole-frame receives. Serial driver version 4.27 with no serial options enabled ttyS00 at 0x03f8 (irq = 4) is a 16550A ttyS01 at 0x02f8 (irq = 3) is a 16550A eth1: Setting Rx mode to 1 addresses. Now I noticed both have "Address 24", but other than that, I see no obvious potential problem, and even the latter isn't necessarily wrong. I don't really know. I guess it's back to Rubini. Any help or suggestions are appreciated. > > Matt > > ======================================================================== > Contributions/Posts To: [EMAIL PROTECTED] > To Unsubscribe: [EMAIL PROTECTED], "unsubscribe" in message body > Report Problems to: [EMAIL PROTECTED] > List archive at: http://www.ssc.com/mailing-lists/
