Coupled task failure

Unified Model
,
1

Hi,

I was running HadGEM3-GC3.1-LL, suite u-cr613, and the model failed in coupled task 23060701T0000Z because my disk quota was exceeded. I cleared disk space and retriggered the task, but this failed again with errors like

???
???!!!???!!!???!!!???!!!???!!! ERROR ???!!!???!!!???!!!???!!!???!!!
? Error code: 4
? Error from routine: STWORK
? Error message: STWORK: Number of fields exceeds reserved headers for unit 28
? Error from processor: 22
? Error number: 37
???

How should I fix the issue? The model ran fine for about 50 years before the disk quota was exceeded. I notice there are several size 0 files - e.g. cylc-run/u-cr613/share/data/History_Data/cr613a.pv2306jun - is the problem related to this?

Best wishes,

Rachel

2

Hi,

Unit 28 is assigned to output file pp2, so (as in the NCAS training exercises) on PUMATest in roses/u-cr613/app/um/rose-app.conf I’ve changed reinit_step from 90 to 60:

file_id=‘pp2’
!!filename=‘$ROSE_DATAC/$RUNID.pp2’
filename_base=‘$DATAM/${RUNID}a.pc%C’
l_reinit=.true.
packing=5
reinit_end=-1
reinit_start=0
reinit_step=60
reinit_unit=2
reserved_headers=32000

I reloaded the suite, and re-triggered coupled task 23060701T0000Z. However, the model now fails with new error:

???
???!!!???!!!???!!!???!!!???!!! ERROR ???!!!???!!!???!!!???!!!???!!!
? Error code: 4307
? Error from routine: CHECK_IOSTAT
? Error message:
? Error reading namelist NLSTCALL_PP
? IoMsg: Incorrect namelist input variable “REINIT_STEP” or mismatched input value.
? Please check input list against code.
? Error from processor: 0
? Error number: 0
???

Do you have any idea about how to fix the original problem?

Best wishes,

Rachel

3

Hi Rachel

I don’t understand why the model decided that Number of fields exceeds reserved headers for unit 28 after running OK for 5 years.

I no longer see the error: IoMsg: Incorrect namelist input variable “REINIT_STEP” or mismatched input value in the output log, but the model has reinit_step=60.

So, I’m not sure what is going on?

Grenville

5

Hi Grenville,

Yesterday, I switched reinit_step back to 90 and retriggered the task so I could see the original problem again. I’m not sure why reinit_step=60 now, maybe I didn’t save it correctly?
If it’s helpful, I can retrigger the task again with reinit_step=60, so you can see the error IoMsg: Incorrect namelist input variable “REINIT_STEP” or mismatched input value.

Best wishes,

Rachel

6

ah, OK – as a work around try increasing the reserved_headers to 50000 (leave reint_step at 90 days)

7

Hi,

Thanks, I just tried that but still get the error message -
STWORK: Number of fields exceeds reserved headers for unit 28.

Best wishes,

Rachel