[CIG-SEISMO] blowup again!

[mingzhao.china]

Hi,

I'm afraid the same blowup problem happen again when running the benchmark case of prem in the EXAMPLE.See in the attachment,some value became infinity.
  Max non-dimensional potential Ufluid in fluid in all slices =  -Infinity
  Max of strain, eps_trace_over_3_crust_mantle = -Infinity
  Max of strain, epsilondev_crust_mantle  = -Infinity

I'm coming back to 5.0.1,until now it never has such problem,but very slowly.When run a case of forword modelling,using 384 cores,each with 1.44GB memory,it takes more than one day(just an estimate,not finish yet).

Cheers
Ming




mingzhao.china
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geodynamics.org/pipermail/cig-seismo/attachments/20120905/3b837ceb/attachment.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: output_solver.txt
Type: application/octet-stream
Size: 14507 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-seismo/attachments/20120905/3b837ceb/attachment.obj