[MITgcm-support] Problems with daily snapshot of MITgcm

Yuan Lian lian at email.arizona.edu
Fri Jun 15 17:51:39 EDT 2007


Hi Martin,

I am running atmospheric model with cubed sphere grids. All STDOUT.* 
stopped at the first time step of the 2-D solver:
cg2d: Sum(rhs),rhsMax =   1.56125112837913E-17  2.40903481122240E-05

I didn't use EXF package.

Thanks!

Yuan

Martin Losch wrote:
> Yuan,
> you need to provide more details. All I can guess form your error 
> messages that you use 12 cpus, and each cpu is stopped with SIGSEGV: 
> 11 (whatever that means). What does STDERR.00[0-11] say, what's the 
> last entry in STDOUT.*? what packages do you use, etc.
>
> There has been a change in the exf-package, which means that all 
> data.exf and data.pkg need to be changed in case you are using, but 
> this is a wild guess given your information.
>
> Martin
>
> On 15 Jun 2007, at 20:58, Yuan Lian wrote:
>
>> Hi,
>>
>> I recently updated MITgcm. I just found my mode won't run with the 
>> most up-to-date MITgcm. The compilation
>> went through but runtime error exists. Here are error messages:
>>
>> p3_27498:  p4_error: interrupt SIGSEGV: 11
>> p4_27500:  p4_error: interrupt SIGSEGV: 11
>> p6_27505:  p4_error: interrupt SIGSEGV: 11
>> p11_27518:  p4_error: interrupt SIGSEGV: 11
>> p9_27513:  p4_error: interrupt SIGSEGV: 11
>> p1_27493:  p4_error: interrupt SIGSEGV: 11
>> p0_27481:  p4_error: interrupt SIGSEGV: 11
>> p10_27515:  p4_error: interrupt SIGSEGV: 11
>> p5_27503:  p4_error: interrupt SIGSEGV: 11
>> p2_27495:  p4_error: interrupt SIGSEGV: 11
>> p7_27508:  p4_error: interrupt SIGSEGV: 11
>> p8_27510:  p4_error: interrupt SIGSEGV: 11
>>
>>
>> The model runs fine with MITgcm_checkpoint58
>>
>> Anyone can give me a hand on this?
>>
>> Thanks,
>> Yuan
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>




More information about the MITgcm-support mailing list