Adam> The output of the previously suggested diagnostics would be Adam> helpful, though it does sounds like the ethernet card is cooked Adam> (but let's keep our fingers crossed). In the meantime, can you Adam> please refresh my memory ... Does the Ultra 1 use the older sbus Adam> for cards, or is a PCI based machine? I may have an sbus Adam> ethernet card laying around. The U1 is an Sbus based system, runs at either 140 or 170Mhz. In this case, I'd do the following to help diagnose things: 1. Halt to the ok prompt. 2. set the diag level up: ok setenv diag-level max ok setenv diag-switch? true ok 3. Then reset the system or try testing it with ok reset or ok test-all and see what it says. You can also test the network with ok setenv tpe-link-test? true ok test net or you might need to do a full reset to get the link to test. This will help narrow down the problem. I *tihnk* I might have an sbus ethernet card somewhere, but it's hard to say. If anything, it might be time to just upgrade to something newer and more supportable. The SunBlade 100s are pretty darn cheap these days, as are the old Netras and other old Sparc hardware. Good luck! John

I have an Ultra 60 that needs a home. I have Aurora on it, though it's an old version. The system has been in storage for two years, at least. If anyone is interested, there's the system, the monitor, mouse, and keyboard up for grabs. I can probably make the January meeting. ...Bill On Mon, Dec 22, 2008 at 10:27 AM, Adam Gomes <adam@unixgeek.net> wrote:

The output of the previously suggested diagnostics would be helpful, though it does sounds like the ethernet card is cooked (but let's keep our fingers crossed). In the meantime, can you please refresh my memory ... Does the Ultra 1 use the older sbus for cards, or is a PCI based machine? I may have an sbus ethernet card laying around.

On Sunday 21 December 2008 18:58:01 Ken Jones wrote:

...

Oops!

I just looked inside my ULTRA SPARC ONE. The ethernet I/O appears to be embedded in the big mother board. That is not good if I am correct.

Any suggestions??

Ken Jones ----- Original Message ----- From: Ken Jones To: wlug@mail.wlug.org Sent: Sunday, December 21, 2008 6:31 PM Subject: Aurora - SPARC question....

Folks,

I've not been WLUG active for a year or more. Back then the half of my 111 gig PC disc which was dual booted with FEDORA developed a permanent error. The other half's XP installation has been OKAY. I have learned to us WINDOWS.

Meanwhile my SUN ULTRA running as an Aurora SPARC SAMBA file server since 1993 has been chugging along just fine.

I live in Fitchburg. We have been without power for nine days. 110 volts came up yesterday evening. Much to my chegrin my SUN was unhappy. I used a rescueCD to boot. All came back except my ethernet card. It would not start. The little light at the far end on the IP cord does not glow...

--------- /etc/rc.d/init.d/network restart

says, "Bringing up interface eth0: Determining IP information for eth0 ... failed. [FAILED]"

The card is claimed to be a 'sunlance' card.

Any ideas? Is the card fried do you think?

I have a second SPARC machine. It's ethernet was working last time it was powered. Must I swap?

Ken Jones

-- Adam Gomes adam@unixgeek.net Public Key @ http://www.thatguygomer.com/pubkey.txt

_______________________________________________ Wlug mailing list Wlug@mail.wlug.org http://mail.wlug.org/mailman/listinfo/wlug

Eric Martin wrote:

Have you tried to ping the loopback address? I know that a fried card won't respond to lo replies (127.0.0.1) in Windows, but I'm not sure what the behavior is in *NIX. That being said, if the loopback interface isn't up then we have our answer.

That trick won't work under any self-respecting 'nix clone. Unix based systems have a virtual interface that gets configured with 127.0.0.1/8, and doesn't care about the state of our physical adapters. The reason that works at all under Windows is because Windows doesn't actually have a loopback interface device driver. (okay, technically there is one, but it's not installed by default.) Instead, there's some special casing in the IP stack that makes it consider 127.0.0.1 a hidden, non-removable IP alias on each regular network interface. As a side note, if you're on the same local network as the victime, you can often leverage this gross hack to bypass Windows firewall rules by sending it packets with the correct destination MAC address and an IP address of 127.0.0.1. -- Frank Sweetser fs at wpi.edu | For every problem, there is a solution that WPI Senior Network Engineer | is simple, elegant, and wrong. - HL Mencken GPG fingerprint = 6174 1257 129E 0D21 D8D4 E8A3 8E39 29E3 E2E8 8CEC

Frank Sweetser wrote:

Eric Martin wrote:

...

Have you tried to ping the loopback address? I know that a fried card won't respond to lo replies (127.0.0.1) in Windows, but I'm not sure what the behavior is in *NIX. That being said, if the loopback interface isn't up then we have our answer.

That trick won't work under any self-respecting 'nix clone. Unix based systems have a virtual interface that gets configured with 127.0.0.1/8, and doesn't care about the state of our physical adapters.

The reason that works at all under Windows is because Windows doesn't actually have a loopback interface device driver. (okay, technically there is one, but it's not installed by default.) Instead, there's some special casing in the IP stack that makes it consider 127.0.0.1 a hidden, non-removable IP alias on each regular network interface.

As a side note, if you're on the same local network as the victime, you can often leverage this gross hack to bypass Windows firewall rules by sending it packets with the correct destination MAC address and an IP address of 127.0.0.1.

Very cool hack! Thanks for the info on NIX and lo, I figured it might be different altogether but I wasn't sure. -- Eric Martin Key fingerprint = D1C4 086E DBB5 C18E 6FDA B215 6A25 7174 A941 3B9F