[MITgcm-support] Restarting the MITgcm

[Uli Riemenschneider]

Problem solved. I set the model to produce the permanent checkpoint 
files, so I don't need to fiddle with the rolling restart files - The 
problem was probably that I miscounted the number of 0's in the file 
name when renaming the rolling checkpoint files, or something silly like 
that...

Thanks for all your suggestions. It is working now!

Best
Uli

Patrick Heimbach wrote:
> Uli,
> 
> if you generate this pickup, that should be fine,
> even for multi-tiled output
> (you probably have writeGlobalFiles=.TRUE. when you
> generated that pickup).
> The mdsreadfield routine looks / tries to find a file
> in following order: (in your case FILENAME='pickup.0000000010')
> FILENAME
> FILEMAME.data
> FILENAME.00?.00?.data
> 
> which means you should be OK in principle.
> You probably did try to set your niter0=10
> to agree with the pickup iteration number, did you?
> 
> Next thing, as Dimitris pointed out,
> check the end of your STDOUT (not STDERR).
> While STDERR has the error message that you reported,
> STDOUT should have the name of the file it couldn't open.
> Can you send us this file name, or maybe the last
> 20 lines, or so, of STDOUT.0000?
> 
> -Patrick
> 
> 
> 
> Quoting Uli Riemenschneider <uriemenschneider at whoi.edu>:
> 
> 
>>I think I see the problem now. I am only creating one pickup file for 
>>each checkpoint which has no extension for the different tiles. So there 
>>is only one pickup.0000000010.data for example without the .001.001 etc. 
>>at the end. What do I need to set to produce these pickupfiles for the 
>>individual tiles??
>>
>>Thanks,
>>Uli
>>
>>
>>>Hi Uli,
>>>
>>>If you're running a multi-tile experiment (which it appears to be), then
>>>there will be a series of files of the form:
>>>
>>>  name.TIMESTEP.TID1.TID2.data
>>>
>>>where TIMESTEP is as Martin described and the TID1.TID2 parts are the
>>>x,y tile number indicies.  Things to check are:
>>>
>>>  1) did you make sure to create the pickup files for all the 
>>>     different needed TID1.TID2 values (001,002,...)?
>>>
>>>  2) did you look in the end of the output (STDOUT.xxx) to see 
>>>     what was happening within the model when it failed to read 
>>>     the file--this should give a good idea
>>>
>>>Ed
>>>
>>
>>
>>-- 
>>*********************************************************
>>Ulrike Riemenschneider,  Postdoctoral Investigator
>>Physical Oceanography Dept. MS #21
>>Woods Hole Oceanographic Institution
>>Woods Hole, MA 02543, USA
>>
>>Phone: (+1) 508 289 2916 Fax: (+1) 508 457 2181
>>*********************************************************
>>
>>_______________________________________________
>>MITgcm-support mailing list
>>MITgcm-support at mitgcm.org
>>http://dev.mitgcm.org/mailman/listinfo/mitgcm-support
>>
> 
> 
> 
> --------------------------------------------------------
> Patrick Heimbach   Massachusetts Institute of Technology
> FON: +1/617/253-5259                  EAPS, Room 54-1518
> FAX: +1/617/253-4464             77 Massachusetts Avenue
> mailto:heimbach at mit.edu               Cambridge MA 02139
> http://www.mit.edu/~heimbach/                        USA
> 
> 


-- 
*********************************************************
Ulrike Riemenschneider,  Postdoctoral Investigator
Physical Oceanography Dept. MS #21
Woods Hole Oceanographic Institution
Woods Hole, MA 02543, USA

Phone: (+1) 508 289 2916 Fax: (+1) 508 457 2181
*********************************************************