CMCI exceptions happened and MCE entry state transition made Xen crashed.

CMCI exceptions happened and MCE entry state transition made Xen crashed.

[flyfan05]

[-- Attachment #1.1: Type: text/plain, Size: 850 bytes --]

Hi all, 
Three days ago, the server reported lots of CMCI exceptions and Xen 3.4.2 printed hundreds of "CMCI: send CMCI to DOM0 through virq" messages to the console .  From the console output, Then I can see that Dom0 try to read the MSR_CAP regs by #GP trap in order  to log the MCA error. 


I am not sure why so many CMCI happened , maybe there were some thing wrong with the hardware.  But unfortunately the server crashed at the end. The Xen BUG ON at
    mctelem_append_processing()   
    ->  MCTE_TRANSITION_STATE(tep, COMMITTED, PROCESSING)  
    ->  BUG_ON(MCTE_STATE(tep) !=    (MCTE_F_STATE_##old)); 
  The output of the console is like this: 
(XEN) Xen bug on at mctelem.c : Line 437


Why the state of the entry is not correct ? Some one change that unexpected?  If any body even resolve this kind problems, Pls do me a favor. 


--Van

[-- Attachment #1.2: Type: text/html, Size: 1341 bytes --]

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

CMCI exceptions happened and MCE entry state transition made Xen crashed.

[flyfan05]

[-- Attachment #1.1: Type: text/plain, Size: 898 bytes --]

Hi all, 
Three days ago, the server reported lots of CMCI exceptions and Xen 3.4.2 printed hundreds of "CMCI: send CMCI to DOM0 through virq" messages to the console .  From the console output, Then I can see that Dom0 try to read the MSR_CAP regs by #GP trap in order  to log the MCA error. 


I am not sure why so many CMCI happened , maybe there were some thing wrong with the hardware.  But unfortunately the server crashed at the end. The Xen BUG ON at
    mctelem_append_processing()   
    ->  MCTE_TRANSITION_STATE(tep, COMMITTED, PROCESSING)  
    ->  BUG_ON(MCTE_STATE(tep) !=    (MCTE_F_STATE_##old)); 
  The output of the console is like this: 
(XEN) Xen bug on at mctelem.c : Line 437


Why the state of the entry is not correct ? Some one change that unexpected?  If any body even resolve this kind problems, Pls do me a favor. 


--Van



网易163/126邮箱百分百兼容iphone ipad邮件收发

[-- Attachment #1.2: Type: text/html, Size: 1566 bytes --]

[-- Attachment #2: Type: text/plain, Size: 137 bytes --]

_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users

Re: CMCI exceptions happened and MCE entry state transition made Xen crashed.

[Keir Fraser]

Sounds like your system probably has a bad memory DIMM. If the MCE logic is
causign problems you can turn it off with mce=0 Xen boot parameter. This
will cause all these correctable errors to be ignored rather than logged;
and uncorrectable errors will cause immediate hypervisor stop-and-crash
rather than have dom0 attempt to fix up.

 K.

On 12/12/2010 07:08, "flyfan05" <flyfan05@163.com> wrote:

> Hi all, 
> Three days ago, the server reported lots of CMCI exceptions and Xen 3.4.2
> printed hundreds of "CMCI: send CMCI to DOM0 through virq" messages to the
> console .  From the console output, Then I can see that Dom0 try to read the
> MSR_CAP regs by #GP trap in order  to log the MCA error.
> 
> I am not sure why so many CMCI happened , maybe there were some thing wrong
> with the hardware.  But unfortunately the server crashed at the end. The Xen
> BUG ON at
>     mctelem_append_processing()
>     ->  MCTE_TRANSITION_STATE(tep, COMMITTED, PROCESSING)
>     ->  BUG_ON(MCTE_STATE(tep) !=    (MCTE_F_STATE_##old));
>   The output of the console is like this:
> (XEN) Xen bug on at mctelem.c : Line 437
> 
> Why the state of the entry is not correct ? Some one change that unexpected?
> If any body even resolve this kind problems, Pls do me a favor.
> 
> --Van
> 
> 
> 网易163/126邮箱百分百兼容iphone ipad邮件收发
> <http://help.163.com/special/007525G0/163mail_guide.html?id=2716>
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel