I just got a Web Power Switch, and hooked it up to the lab machines in the basement. I can now reset them with a web page… which is going to get scripted really soon.

I am getting [sick]{20060805101941} [of]{20060803233234} [this]{20060802210126}. It froze again.

So to date I have disabled swapped out the motherboard (but still have the same NForce4 chipset), enabled all sorts of debug features in the kernek, removed the Nforce4 network card (based on a hint on a web forum), and put in a PCI NIC. It’s still unstable.

I had noticed that it never froze under console, just under X. The next attempt is to disable DRM. I did this through the xorg.conf… let’s see what happens next. I am no longer optimistic.

Over the last [few]{20060802210126} [days]{20060803233234}, I have been experiencing odd crashes. I have been able to narrow it down to network activity. Most of the time this has been triggered with NFS access; I assume that it has something to do with large UDP fragments; my MTU is 1500.

The last couple of days I have had a serial console connected to the machine, and my laptop standing by waiting for another crash. It occurred, but something really bad happened because on the console I only got:

[Yesterday]{20060802210126}, I wrote about how my Opteron 170 was crashing at random times. Today I have a working system.

My new shiny Opteron 170 just came in. I used to say that Opteron 1xx line was a waste of money because it was basically the Athlon64 but more expensive because of the Opteron branding. Recently AMD made the choice easier by dropping dual-core Athlon64 processors with 2M of L2 cache. The last line of Athlon64 S939 will have a 512k L2 cache. That’s a mere 256k per core. And isn’t that only twice the size of the L1 cache? yuck!

So, I blew the extra $300 on the extra 1.5M of L2 cache.