Changes between Version 1 and Version 2 of user/flavoni/libIGCM_v3.6/list_of_possible_errors


Ignore:
Timestamp:
2015-10-30T22:17:25+01:00 (5 years ago)
Author:
flavoni
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • user/flavoni/libIGCM_v3.6/list_of_possible_errors

    v1 v2  
    44 * '''To perform a second simulation after a run with !PeriodState=Fatal''' 
    55 
    6   rm Job_name/run.card;  rm Script_Output* 
     6  rm Job_name/run.card;  rm Script_Output* 
     7  rm everything created during the simulation in IGCM_OUT 
    78 
    8   but do not cancel run.card.'''init''' file, if not run.card file is not re-created. 
     9  but do not cancel run.card.''' init''' file, if not run.card file is not re-created. 
    910 
    10   ( File run.card.init is created with ins_job script ) 
     11  ( File run.card.init is created with ins_job script )[[BR]][[BR]]Submit the main job 
     12  --- 
     13  OR: cd EXPERIENCE[[BR]]../../../libIGCM/'''clean_month.job''' 
     14 
     15 * '''How to continue a simulation ''' 
     16 
     17  Update config.card with the new !DateEnd. Do not change !DateBegin.''' ''' 
     18  Change !PeriodState in run.card to !OnQueue (or let clean_month job do it for you)''' ''' 
     19  Submit the main job 
     20 
     21 * ''' How to continue a simulation that crashed previously''' 
     22 
     23  If you want to relaunch a crashed simulation you need first to remove all files produced for the crashed period. 
     24  Use the script modipsl/libIGCM/clean_month.job and answer yes to delete files and change run.card 
     25  Submit the main job 
    1126 
    1227 * '''!RebuildFromArchive=NONE''' 
    1328 
    14   In MPI it is good to use Rebuild in "archive" machine, to not ovrerload calcul machine. [[BR]]  You have to put in config.card in POST Session: 
     29  In MPI it is good to use Rebuild in "archive" machine, to not ovrerload calcul machine. [[BR]]   You have to put in config.card in POST Session: 
    1530 
    16   !RebuildFrequency  =1Y [[BR]]  !RebuildFromArchive  =true 
     31  !RebuildFrequency   =1Y [[BR]]   !RebuildFromArchive   =true 
    1732 
    18   NOTA BENE: If you put !RebuildFrequency  =1Y and you forget to set !RebuildFromArchive  =true, it is very dangerous. NEVER DO IT! 
     33  NOTA BENE: If you put !RebuildFrequency   =1Y and you forget to set !RebuildFromArchive   =true, it is very dangerous. NEVER DO IT! 
    1934 
    2035 * '''How to do a long job without exit from queue''' 
    2136 
    22   You can play with Duration Experience, PeriodLenght   and PeriodNb  , elapstim_req in this way :[[BR]]  in '''  config.card'''   Duration Experience is = !DateBegin   - !DateEnd   [[BR]]  in '''  config.card'''   !PeriodLenght   is how many time you do restart files[[BR]]  in '''  Job_Name'''   !PeriodNb   is  in '''  Job_Name'''   elapstim_req is elapsed time limit 
     37  You can play with Duration Experience, PeriodLenght    and PeriodNb   , elapstim_req in this way :[[BR]]   in '''   config.card'''    Duration Experience is = !DateBegin    - !DateEnd    [[BR]]   in '''   config.card'''    !PeriodLenght    is how many time you do restart files[[BR]]   in '''   Job_Name'''    !PeriodNb    is  in '''   Job_Name'''    elapstim_req is elapsed time limit 
    2338 
    24         Your job is out from queue every time that is finished a PeriodNb  , so you can do : 
     39  Your job is out from queue every time that is finished a PeriodNb   , so you can do : 
    2540 
    26   !PeriodNb   = ( Duration Experience ) / !PeriodLenght 
     41  !PeriodNb    = ( Duration Experience ) / !PeriodLenght 
    2742 
    2843  but be careful with elapstim_req, because this is a limit, you've to be sure that it exists a queue that can hold you.