Highlighted
STLTHX
Bronze

Random Reboots with a R710 and Windows 2008 R2

I have a brand new server implementation with a R710 and Windows Server 2008 R2.

The server has been through two unexplained and random reboots so far.

There has been no blue screen of death and no bugchecks.

This is a simple server installation who's role is simply Hyper-V v2 and that's it.

Is anyone else experiencing this issue? All firmware is at the latest.

How does one go about troubleshooting a problem such as this?

Of course, since Dell OpenManage 6.2 isn't out yet I am unable to perform any sort of diagnostics or log examination.

Thanks.

0 Kudos
12 Replies
STLTHX
Bronze

Re: Random Reboots with a R710 and Windows 2008 R2

I just found a possible reason and resolution.

Is anyone else experiencing this same problem?

http://blogs.msdn.com/virtual_pc_guy/archive/2009/10/16/hyper-v-hotfix-for-0x00000101-clock-watchdog-timeout-on-nehalem-systems.aspx

0 Kudos

Re: Random Reboots with a R710 and Windows 2008 R2

We have the same problem. 2 R710 with Windows 2008 R2, Hypervisor-Role and random reboots if server is under heavy load. I will try the hotfix.

0 Kudos
STLTHX
Bronze

Re: Random Reboots with a R710 and Windows 2008 R2

Looks like I spoke too soon.

After disabling the C1E state in the server's BIOS our Dell R710 running Hyper-V R2 on 2008 R2 bluescreened last night.

It was about 7 weeks since the last bluescreen due to this bug.

Is anyone else still seeing bugchecks because of this problem even after disabling C1E in the BIOS?

0 Kudos
mbenadiba
Copper

Re: Random Reboots with a R710 and Windows 2008 R2

I am also experiencing this exact same issue.  We have two identically configured PE R 710 servers. Only one is exhibiting the signs of this issue while the other is not.  Dell Tech Support confirmed we should disable C1-E in the BIOS.  The only noticeable change is, with C1-E disabled, the server just completely freezes, including from DRAC, and the only way to bring it back is a full power cycle manually (in our case via DRAC).  By leaving the C1-E state enabled, the server still crashes but it Blue Screens and automatically reboots itself which is much more useful than it halting.

http://www.mbccs.com

0 Kudos

Re: Random Reboots with a R710 and Windows 2008 R2

if time permits, my first test would be to swap all the memory between the systems, see if the problem follows the memory. you could also swap the hard drives between systems as well (be careful when importing the discs between systems, if possible, backup the systems first).

0 Kudos

Re: Random Reboots with a R710 and Windows 2008 R2

For R710 system reboot- Not sure what type of load is in the server but try the patch # KB975530 and see if you still see the behavior. 

For Processor C1E state – Try the patch # KB974090.

Thanks

0 Kudos
STLTHX
Bronze

Re: Random Reboots with a R710 and Windows 2008 R2

It turns out that I had to end-up disabilng all of the C-States in the BIOS.

Even after disabling C1E I was still getting BSOD's.

0 Kudos

Re: Random Reboots with a R710 and Windows 2008 R2

This is a better explanation of the issue and possible work arounds
http://support.microsoft.com/kb/975530

0 Kudos
STLTHX
Bronze

Re: Random Reboots with a R710 and Windows 2008 R2

REYBEAST1 brings up a good point.

In addition to disabling C1E and all of the otehr C states make sure to install this KB hotfix.

After doing the above my BSOD's disappeared altogether.

0 Kudos