If you saw my Christmas wishlist over at NewEgg this year, you probably realized I was in the market to build a new computer. Last Tuesday I finally ordered all the parts I needed, which, thanks to my lovely wife and the brother-in-law, was a shorter list than it began with.
The shopping list:
- Shuttle SD37P2
- Intel Core2 Duo E6400 (2.13 GHz, 1066 MHz FSB, 2 MB L2 cache)
- 2 Samsung SpinPoint SP2504C 250GB SATA disks
- NEC 18x DVD burner
- 2 2GB (2x1GB) DDR2 SDRAM 667 GeIL chips, with heat spreaders
Yeah, thats way too much RAM for a desktop or home Windows Server 2003 install. The intent is to install CentOS 4.4 x86_64 and VMware Server and virtualize most of the other servers in my home office. My aging, and probably on its last legs, PIII-733 megatower “server” is in desperate need of new hardware, and this is where it will be going.
Issues along the way so far:
- I seem to have the v2 version of the SD37P2, as I’m running a BIOS that only claims to work on the v2. Not a big deal, but I’m not sure what they “fixed” with this version.
- The motherboard, or the Core2 Duo chip itself, won’t run the stock CentOS 4.4 SMP kernel with ACPI enabled. Not a big deal, but it would be nice if the front power button and shutdown -h now actually worked. I’m hoping RedHat gets RHEL5 out the door soon and CentOS isn’t far behind. Or maybe a BIOS update will fix this. For others searching, it causes a “kernel panic: Oops” or another odd kernel panic, and the solution is adding acpi=off to the end of your kernel line in grub.conf.
- The Intel 82572EI Pro/1000 PCI Express ether card that I bought a while back isn’t supported out of the box. I had to go download and recompile a driver from their support site, but the instructions were pretty good
- My APC SmartUPS 700 can’t take both Shuttle boxes (SB81P is the other) hanging off of it, but I suppose I need to get a bigger UPS anyway. Or once the megatower is gone, I can re-distribute the load.
I suppose these are all the pains of being out near the bleeding edge of hardware, at least for home Linux systems. For now, I’m going to let the system just sit and burn in, the processor and fans need a good workout before I’ll call them ready for work.
And, on a side note, the tracking info was kind of entertaining:
12/29/2006 17:35:00 DELIVERED[D] CHAMPAIGN, IL, US PORCH
12/29/2006 05:22:00 OUT FOR DELIVERY[I] URBANA, IL, US
12/29/2006 04:50:00 ARRIVAL SCAN[I] URBANA, IL, US
12/29/2006 02:47:00 DEPARTURE SCAN[I] DECATUR, IL, US
12/29/2006 00:22:00 LOCATION SCAN[I] DECATUR, IL, US
12/29/2006 00:10:00 ARRIVAL SCAN[I] DECATUR, IL, US
12/28/2006 20:24:00 DEPARTURE SCAN[I] DAVENPORT, IA, US
12/28/2006 18:45:00 ARRIVAL SCAN[I] DAVENPORT, IA, US
12/28/2006 15:49:00 DEPARTURE SCAN[I] DES MOINES, IA, US
12/28/2006 14:52:00 LOCATION SCAN[I] DES MOINES, IA, US
12/28/2006 13:20:00 UNLOAD SCAN[I] DES MOINES, IA, US
12/28/2006 13:16:00 ARRIVAL SCAN[I] DES MOINES, IA, US
12/28/2006 08:39:00 DEPARTURE SCAN[I] SPOKANE, WA, US
12/28/2006 05:12:00 ARRIVAL SCAN[I] SPOKANE, WA, US
12/28/2006 04:30:00 ADVERSE WEATHER CONDITIONS CAUSED THIS DELAY[X] PORTLAND, OR, US
12/28/2006 02:07:00 DEPARTURE SCAN[I] ONTARIO, CA, US
12/28/2006 00:40:00 LOCATION SCAN[I] ONTARIO, CA, US
12/27/2006 22:17:00 ARRIVAL SCAN[I] ONTARIO, CA, US
12/27/2006 21:38:00 DEPARTURE SCAN[I] BALDWIN PARK, CA, US
12/27/2006 21:36:00 ARRIVAL SCAN[I] BALDWIN PARK, CA, US
12/27/2006 15:43:00 ORIGIN SCAN[I] BALDWIN PARK, CA, US
12/27/2006 20:01:15 BILLING INFORMATION RECEIVED[M] BALDWIN PARK, CA, US
I don’t think I’ve seen “adverse weather conditions” before in a tracking report.