[CP2K-user] segmentation fault core dumped (MD run stops)

[Javad Shirani]

Dear CP2K users,

I am running cp2k/4.1 on Cedar for long MD runs of solvated-ion and liquid 
water systems. The issue I am facing is that although we set the 
calculations to run for 12 hours, they start running fine and after ~1.5 
hours it is crashed by "segmentation fault error". The error seems not to 
be related to convergence of SCF loop.

I use cp2k.popt to run the calculation on 4 nodes and 192 cpus. Changing 
the number of nodes or cpus did not help improving the result.

Please find attached my input files and job submission files for both cases 
of liquid water and solvated-Fe systems.

I appreciate any comment which can help getting ride of this error.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20201029/1b8eb5a2/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: input_output.zip
Type: application/x-zip
Size: 804327 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20201029/1b8eb5a2/attachment.bin>