Windows 2003 Domain Controller restarting multiple times a day unexpectedly

can you post you memory dumb file ?

Where would I find the file?

follow this steps to create or find the files

https://www.experts-exchange.com/articles/Programming/Languages/Scripting/DOS_Windows/How-To-Troubleshoot-BSOD-Blue-Screen-Of-Death-and-Windows-Stop-Errors.html

http://support.microsoft.com/kb/307973

They are usually located in %systemroot%/minidump (%systemroot%/minidump).

It is set to write a kernel dump to %systemroot%\MEMORY.DMP, however when looking in that directory or doing a search, for *.DMP on the hard drive, nothing is there.

hmm then its is not crashing it is manually shutting down or restarting , it must be a software which causing  that . can you check you event viewr for reboot or system errors on both application and system logs and post any erros ?

There's NOTHING running on this domain controller besides DNS services......And the reboots are pretty random.  Here's a screenshot of the log:



Picture-1.png

Where it is highlighted is when I start seeing "Services started", etc....

Do you have an IPMI connection to it, is it an HP server or IBM. HP calls is integrated lights out can't remember how does IBM name it. It could be a hardware issue -Power supply unit? It would help if you could get the IPMI logs as well.

Just noticed after checking the logs, there are quite a few 3011 and 3012 errors.  I see that I can rebuild the Performance Index logs by following the instructions in this article:
http://support.microsoft.com/kb/300956

1.) Is there a good chance that these errors :
Unloading the performance counter strings for service WmiApRpl (WmiApRpl) failed. The Error code is the first DWORD in Data section.
could be causing the reboots?

2.) Is it pretty safe to rebuild the performance counters on a domain controller (my PDC emulator)?

Anyone?  
Even these errors show up AFTER the reboots, I'm thinking they must be related somehow.  Should I try to rebuild the performance counters?

I've had servers randomly reboot and when there is nothing useful in the logs, it is almost always a hardware problem.  I'd download the latest manufacturer's offline diagnostics and run them.  

I recommend the offline over the online diagnostics only because the last time this happened, HP's online found nothing, while the offline found it to be a memory/motherboard issue.

Good luck.

I would second that. What kind of hardware is it, is it HP or IBM server. Or Supermicro? No name?

What kind of a power supply does it have?  

Can you try replacing with a another one, just to exclude that possibility?

Did you recently add any hardware to it?

We are a Dell shop.  It is a Poweredge 860 with a single power supply.  There is absolutely NOTHING in the logs BEFORE the reboots, just a gaggle of events afterwards that always starts with the "The system shutdown on ----- was unexpected", and then all the services starting, etc....I will try to track down the Dell offline diagnostics.

Why not try running some online diagnostics.

http://support.euro.dell.com/support/downloads/driverslist.aspx?c=si&l=en&s=gen&ServiceTag=&SystemID=PWE_P4_860&os=WNET&osl=sl&catid=&impid=

And see if anything comes up..

Everything comes up clean.  I've been doing some research and found some other folks on the Dell forum that had the same issue with other PE models.  Most of them reloaded the OS and were good to go, however this REALLY makes me nervous, as the server in question is my MAIN DC (First in the organization) and holds some FSMO roles, DNS and DHCP.......How easy/hard would it be for me to "reload" the OS and what is involved?

Dell replaced the motherboard yesterday late afternoon.  It then rebooted 3 times overnight.  Dell came out again this morning and replaced the CPU, Controller board and Power Supply.  So far so good.  Is there any way to have an alert sent via email if/when the machine reboots itself again?