Negative mass in set_thermodynamic Error

Unified Model
,
1

Hi,

I am running the nesting suite u-eb261 on Monsoon3, and the following task has failed:

20190103T0000Z/Europe_km11_RAL3P3_um_fcst_000

The error message says:

Error code: 100

Error from routine: set_thermodynamic

Error message: A total of 2 points had negative mass in set_thermodynamic. This indicates the pressure fields are inconsistent between different levels and the model is about to fail.

Error from processor: 122

Error number: 55

I have another run using the same nesting-suite setup which did not encounter this error and completed successfully. The only difference between the two runs is the aerosol loading.

I’m not sure what has caused this error. Could you advise what might have caused this or any diagnostics I should run?

Many thanks for your help.

Best,

Kexin

2

Hi Kexin.

Is this error reproducible? I.e. do you get exactly the same error at the same place if you run it again?

Cheers

Jonny

3

Hi Jonny,

Yes, it is reproducible. I get it every time when the task is re-tried.

Thanks,

Kexin

4

Thanks Kexin

I’ve started a test run from the head of the trunk of u-eb261 (see screenshot), I’ll see if I get any errors.

Couple of questions:

  1. Is it the expected behaviour to lauch all the active cycle points simultaneously? As you can see from the screenshot, the cycle points for January 1, 2 and 3 are launched when the workflow starts.
  2. If you have any local commits to u-eb261 please commit them using fcm commit since I can’t see your $HOME directory.
  3. Is there a particular reason why you are using a Cylc 7 setup (i.e. with a suite.rc file)? Currently Cylc 7 workflows can be run using Cylc 8 but not for long…
$ cylc val .
WARNING - Backward compatibility mode ON - support for suite.rc files will be removed at 8.7.0
Valid for cylc-8.6.5

Cheers

Jonny

image
image1401×715 59 KB