[ET Trac] [Einstein Toolkit] #797: N13CarpetRegrid211snap_coarseE does not hold after being enforced in cell centering

Einstein Toolkit trac-noreply at einsteintoolkit.org
Tue Apr 10 12:31:41 CDT 2012


#797: N13CarpetRegrid211snap_coarseE does not hold after being enforced in cell
centering
--------------------+-------------------------------------------------------
 Reporter:  rhaas   |       Owner:  eschnett
     Type:  defect  |      Status:  new     
 Priority:  major   |   Milestone:          
Component:  Carpet  |     Version:          
 Keywords:          |  
--------------------+-------------------------------------------------------
 I have a run (on Caltech's shc and on lonestar) which runs fine for about
 300M or so and then fails with a property enforcement error inside of
 CarpetRegrid.

 Attached please find stdout with Carpet::veryverbose=yes,
 CarpetRegrid2::veryverbose=yes as well as the parameter file.

 It takes about 1hr to reproduce this starting from the checkpoints on shc.
 Most of this time is IO (the checkpoint is iteration 133120 failure occurs
 at preregrid in iteration 133185).

 Turning off cell centering makes the error go away. Turning on refluxing
 makes the run fails (much) earlier with a boxes not contained error (re-
 running this on shc right now, the XSEDE clusters are just too slow).

 In case you have access to shc (or zwicky) and would look at the actual
 output files, the path is:
 /panfs/ds06/sxs/rhaas/cactus/simulations/noreflux/output-0003

-- 
Ticket URL: <https://trac.einsteintoolkit.org/ticket/797>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit


More information about the Trac mailing list