Hello All. First timer here. Here's what's going on..
We have built a new server for our client and it's running fine with one CPU. As soon as we install the second CPU, the system starts randomly rebooting. No BSOD, no apparent errors in the Event Viewer, no dump files getting generated. The screen goes black and then starts the regular boot process. Computer can reboot anywhere from 5 minutes to about 1 hour after it started. With one CPU running it is stable for days.
Here are some specs on the system:
Chassis: Supermicro SC745TQ-R800B with 800W Redundant Power Supply
MBoard: Intel S5000VSA (for multi-core Intel Xeon processors)
BIOS: Version - S5000.86B.10.00.0088.03142
0081550 / Date - 03/14/2008 (should be the latest)
CPU: Intel Xeon E5405 @ 2.00GHz Quad Core (currently one installed. need to install 2 identical)
Memory: 2 gigs (2 sticks) of Kingston DDR2 (KVR667D2D8F5/1G)
Network: Using 2 onboard NICs and 2 Intel PRO/1000 PCIe x1 NICs
Video: Just integrated onboard video, no separate video card
RAID: HighPoint RocketRAID 2320 8 Channel PCI-e SATA II RAID Controller
Drives: One dedicated 320G SATA system drive and 8 SATA drives connected to the controller card
OS: Windows XP Pro SP3 with all the updates
- In BIOS, with 2 processors installed, when I disabled all of the CPU settings (SpeedStep Technology, Deep C-State Support, Core MultiProcessing, Execute Disable Bit, Hardware Prefetcher, Adjucent Cache Line Prefetch) the system/computer was stable for about 4/5 hours and then started freezing up, then later rebooting again. Removing the second CPU fixed the freezing/rebooting problem right away.
- Again in BIOS, disabling the IO Acceleration under PCI Config as suggested by Microsoft made computer reboot even more often, about every 2 minutes.
- I put the CPU that normally went into slot 2 in place of the first one and it's working just fine, no problem there.
- Updated BIOS, didn't help.
- Under Startup and Recovery settings unchecked the Automatically restart check box and verified that the other 2 are checked and that the small memory dump is going to %SystemRoot%\Minidump. Did not find the minidump folder in the above directory, but rather in c:\windows\minidump (maybe that's where it should be?). There one .dmp file there dated almost a month ago, but nothing new. Debagged that file with WinDbg and it pointed me to the rr323x.sys driver, which is the RocketRAID driver.
****
Unable to load image rr232x.sys, Win32 error 0n2
BugCheck 100000D1, {0, 7, 8, 0}
Probably caused by : rr232x.sys ( rr232x+f3ce )
****
Checked the RAID controller driver, the latest available from the manufacturer. Flashed the RAID card BIOS to the lates version (1.7). Shut down, put in the second CPU - computer rebooted the same way as before after about 50 minutes to an hour. There was again no BSOD or any other errors on the screen. It just went black and started booting up. I can't find anything useful in the system event viewer. No errors there about anything crashing, or other problems. No new .dmp files anywhere on the c:\. CPU temperatures I think OK: right now, on 1 CPU, on core #0 it is showing 141F. All other cores are lower. When I had 2 CPUs installed it showed about the same temp. for core #2 (140F) on second CPU. The rest were lower again.
So that's the story. Any and all help is greatly appreciated. The only thing left that I can think of is swapping the motherboard. Thanks for your time and help. SIBR
Start Free Trial
