[MITgcm-support] MITgcm-support Digest, Vol 198, Issue 11
Martin Losch
Martin.Losch at awi.de
Fri Dec 20 14:48:23 EST 2019
Hi Nadine,
no, doesn’t make sense to me, unfortunately.
Is you configuration small enough that you can post it somewhere (code and input, or maybe all STDOUT.* and STDERR.* files?
Martin
> On 20. Dec 2019, at 19:33, Nadine Steiger <Nadine.Steiger at uib.no> wrote:
>
> Thanks Martin,
>
> I am not entirely sure about how to figure out what causes the error.
> Here is a bit more information on the two different runs that I tried:
>
> When I run this first version
> OB_Jnorth= 1000*-1,
> OB_Jsouth= 1000*1,
> OB_Ieast = 480*-1,
> OB_Iwest = 480*1,
>
> the error message in STDERR is:
> (PID.TID 0100.0001) *** ERROR *** OBCS_CHECK: 10 errors in tile OB set-up
> (PID.TID 0100.0001) *** ERROR *** S/R ALL_PROC_DIE: ending the run
>
> and the last lines in STDOUT.0100 are:
> (PID.TID 0100.0001) OBCS_CHECK: #define ALLOW_OBCS
> (PID.TID 0100.0001) OBCS_CHECK: start summary:
> (PID.TID 0100.0001) useOBCSprescribe = /* prescribe OB values */
> (PID.TID 0100.0001) T
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) useOBCSbalance = /* balance the flow through OB */
> (PID.TID 0100.0001) F
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) OBCS_uvApplyFac = /* Factor to apply to U,V 2nd column/row */
> (PID.TID 0100.0001) 1.000000000000000E+00
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) OBCS_u1_adv_T = /* Temp uses upwind adv-scheme @ OB */
> (PID.TID 0100.0001) 0
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) OBCS_u1_adv_S = /* Salt uses upwind adv-scheme @ OB */
> (PID.TID 0100.0001) 0
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) OBCS_monitorFreq = /* monitor output frequency [s] */
> (PID.TID 0100.0001) 1.200000000000000E+03
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) OBCS_monSelect = /* select group of variables to monitor */
> (PID.TID 0100.0001) 1
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) useOBCStides = /* apply tidal forcing through OB */
> (PID.TID 0100.0001) F
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) tidalPeriod = /* (s) */
> (PID.TID 0100.0001) 10 @ 0.000000000000000E+00 /* I = 1: 10 */
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) OB_indexNone = /* null value for OB index (i.e. no OB) */
> (PID.TID 0100.0001) -99
> (PID.TID 0100.0001) ;
> (PID.TID 0100.0001) ======== Tile bi= 1 , bj= 1 ========
> (PID.TID 0100.0001) OB_Jn = /* Northern OB local indices */
> (PID.TID 0100.0001) 56 @ -99 /* I = -2: 53 */
> (PID.TID 0100.0001) OB_Js = /* Southern OB local indices */
> (PID.TID 0100.0001) 56 @ 1 /* I = -2: 53 */
> (PID.TID 0100.0001) OB_Ie = /* Eastern OB local indices */
> (PID.TID 0100.0001) 54 @ -99 /* J = -2: 51 */
> (PID.TID 0100.0001) OB_Iw = /* Western OB local indices */
> (PID.TID 0100.0001) 54 @ -99 /* J = -2: 51 */
> (PID.TID 0100.0001) OBCS_CHECK: end summary.
>
>
>
>
> When I run the second version with
> OB_Jnorth = 1*0,719*-1,280*0,
> OB_Jsouth = 1*0,719*1,280*0,
> # OB_Ieast = 480*-1,
> # OB_Iwest = 480*1,
>
> the error message is:
> (PID.TID 0100.0001) *** ERROR *** OBCS_CHECK: 78 errors in OB location vs Mask
> (PID.TID 0100.0001) *** ERROR *** S/R ALL_PROC_DIE: ending the run
>
> and the last lines in STDOUT are:
> (PID.TID 0100.0001) ======== Tile bi= 1 , bj= 1 ========
> (PID.TID 0100.0001) OB_Jn = /* Northern OB local indices */
> (PID.TID 0100.0001) 56 @ -99 /* I = -2: 53 */
> (PID.TID 0100.0001) OB_Js = /* Southern OB local indices */
> (PID.TID 0100.0001) 56 @ 1 /* I = -2: 53 */
> (PID.TID 0100.0001) OB_Ie = /* Eastern OB local indices */
> (PID.TID 0100.0001) 54 @ -99 /* J = -2: 51 */
> (PID.TID 0100.0001) OB_Iw = /* Western OB local indices */
> (PID.TID 0100.0001) 54 @ -99 /* J = -2: 51 */
> (PID.TID 0100.0001) OBCS_CHECK: end summary.
> (PID.TID 0100.0001) OBCS_CHECK: #define ALLOW_ORLANSKI
> (PID.TID 0100.0001) OBCS_CHECK: #define ALLOW_OBCS_STEVENS
> (PID.TID 0100.0001) OBCS_CHECK: set-up OK
>
>
> Is this any more helpful?
>
> Thank you,
> Nadine
>
>
>
>
>
> On 20/12/2019 18.00, mitgcm-support-request at mitgcm.org wrote:
>> Send MITgcm-support mailing list submissions to
>> mitgcm-support at mitgcm.org
>> To subscribe or unsubscribe via the World Wide Web, visit
>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>> or, via email, send a message with subject or body 'help' to
>> mitgcm-support-request at mitgcm.org
>> You can reach the person managing the list at
>> mitgcm-support-owner at mitgcm.org
>> When replying, please edit your Subject line so it is more specific
>> than "Re: Contents of MITgcm-support digest..."
>> Today's Topics:
>> 1. Re: OBCS_CHECK (Martin Losch)
>> ----------------------------------------------------------------------
>> Message: 1
>> Date: Fri, 20 Dec 2019 16:40:25 +0100
>> From: Martin Losch <Martin.Losch at awi.de>
>> To: MITgcm Support <mitgcm-support at mitgcm.org>
>> Subject: Re: [MITgcm-support] OBCS_CHECK
>> Message-ID: <E6C4C3DC-3C03-4ECF-8449-3C567ACFAEB0 at awi.de>
>> Content-Type: text/plain; charset="utf-8"
>> Hi Nadine,
>> in this case it makes sense to have a closer look what exactly causes the error. In obcs_check.F (and also in your STDERR.* files) you?ll find that this error is caused by a conflict between CPP-flags and the actual domain specification. You?ll have to look for something like
>> OBCS_CHECK: #undef ALLOW_OBCS_NORTH and
>> OBCS_CHECK: tile bi,bj=',bi,bj, ' has Northern OB
>> That is likely to help you find the problem. Without this information it?s difficult to help.
>> Martin
>>> On 20. Dec 2019, at 16:30, Nadine Steiger <Nadine.Steiger at uib.no> wrote:
>>>
>>> Hei,
>>>
>>> I am trying to change my bed topography of a so far working setup, but get this error message:
>>>
>>> (PID.TID 0100.0001) *** ERROR *** OBCS_CHECK: 10 errors in tile OB set-up
>>> (PID.TID 0100.0001) *** ERROR *** S/R ALL_PROC_DIE: ending the run
>>>
>>> My data.obcs looks as follows:
>>> # Open-boundaries
>>> &OBCS_PARM01
>>> # This flag turns off checking and fixing problematic topography across
>>> # open boundaries.
>>> OBCSfixTopo=.FALSE.,
>>> OB_Jnorth= 1000*-1,
>>> OB_Jsouth= 1000*1,
>>> OB_Ieast = 480*-1,
>>> OB_Iwest = 480*1,
>>> useOBCSprescribe = .TRUE.,
>>> OBNvFile = 'OB_NorthV.bin',
>>> OBSvFile = 'OB_SouthV.bin',
>>> OBCS_monitorFreq= 1200.,
>>> OBCS_monSelect = 1,
>>> &
>>>
>>> I found an earlier email thread where it was suggested to describe the open boundaries from reading in hFacW/S
>>> (http://mailman.mitgcm.org/pipermail/mitgcm-support/2014-September/009468.html)
>>>
>>> In this thread someone writes that:
>>> in fortran, the array sections are accessed (after reading in), as
>>>> OBE = int(hFacW(NNX ,1:NNY,1))
>>>> OBW = int(hFacW(2 ,1:NNY,1))
>>>> OBN = int(hFacS(1:NNX,NNY ,1))
>>>> OBS = int(hFacS(1:NNX,2 ,1))
>>>
>>>
>>> So, I retried it, this time prescribing the boundaries as
>>> OB_Jnorth = 1*0,719*-1,280*0,
>>> OB_Jsouth = 1*0,719*1,280*0,
>>> # OB_Ieast = 480*-1,
>>> # OB_Iwest = 480*1,
>>> where I got the values from hFacS.
>>>
>>> In addition, I put a wall at the western and eastern boundary so i wrote in OBCS_OPTIONS.h:
>>> #undef ALLOW_OBCS_EAST
>>> #undef ALLOW_OBCS_WEST
>>> and commented OB_Ieast and OB_Iwest in data.obcs.
>>>
>>> Even with this more detailed description of OB_Jnorth/south, I still get the same error message.
>>>
>>> What strikes me is, that it worked before and I really only changed the bed topography (which is very idealized) slightly.
>>> The new topography works when I turn off the ice shelf package, the old topography worked with the ice shelf package.
>>>
>>> What am I missing here?
>>>
>>> Cheers,
>>> Nadine
>>>
>>>
>>>
>>> _______________________________________________
>>> MITgcm-support mailing list
>>> MITgcm-support at mitgcm.org
>>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>> ------------------------------
>> Subject: Digest Footer
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
>> ------------------------------
>> End of MITgcm-support Digest, Vol 198, Issue 11
>> ***********************************************
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
More information about the MITgcm-support
mailing list