Host reboots, VMs down, vmkwarning TSC: 452832 cpu0:0

[apilz]

Hello,

does somebody already know how to solve our problem?

One of our hosts reboots couple of times a day. After fixing a hardware issue we still seeing the same problem.

vmkwarning includes the following entries:

Jan  4 01:14:31 ESX02 vmkernel: TSC: 4573792 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 01:18:49 ESX02 vmkernel: TSC: 4516080 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 01:56:21 ESX02 vmkernel: TSC: 4565944 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 02:11:59 ESX02 vmkernel: TSC: 4526448 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 02:31:04 ESX02 vmkernel: TSC: 4525904 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 04:57:52 ESX02 vmkernel: TSC: 4573736 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 05:07:12 ESX02 vmkernel: TSC: 4570424 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 05:24:24 ESX02 vmkernel: TSC: 4518072 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60
Jan  4 13:36:01 ESX02 vmkernel: TSC: 4528320 cpu0:0)WARNING: ACPI: 1192: MCFG@: Table length 64 > 60

Would be great getting your replies soon....

Best regards,

Axel

[kjb007]

What ESX 4.0 update are you on?  This article is probably relevant to you:  http://kb.vmware.com/kb/1018108

-KjB

[apilz]

Hi Kanu,

thank you for fast reply. We are on 4.0.0, 162856, no update yet.

There are two hosts showing 100% match of critical patches. One works w/o any problem, 2nd shows up with the problem. If it will be helpfull for you I am able to upload all relevant support data.

Best regards,

Axel 

[kjb007]

Sounds like you may still have a hardware problem.  Are you able to perform hardware diagnostics against this server?

Are you using any agents?  Have you looked at other logfiles, messages around the time the server reboots?

-KjB

[apilz]

Yes of course we did any kind of hardware diagnostics. Result: Fujitsu changed the RAID HBA.

What kind of agents do you think of (UPS)? If you have any suggestions which log/message we should correlate to the warning message I will appreciate your answer.

Axel

[idle-jam]

if it's identical and then one having problem i believe it would be the likes of bad CPU cores or Memory Dimms. From my experience the diagnostic software will at times missed it. You will need to run the diagnostic file and have the support review thru them. They're pretty accurate .. atleast for me few cases.

[apilz]

We have found a solution by changing the FC HBA. Thanks guys for late reply.