clear_shared

View page source

Clear at_cascade Shared Memory

Prototype

# at_cascade.clear_shared
def clear_shared(all_node_database, job_name) :
   assert type(all_node_database) == str
   assert type(job_name) == str

Purpose

This routine clears the at_cascade shared memory that is in use and has a particular prefix. For example, you may get the following error message when trying to run the cascade:

FileExistsError: [Errno 17] File exists: name

where name ends with _number_cpu_inuse or _job_status. This may happen if the previous fit_parallel did not terminate cleanly; e.g., if the system crashed.

all_node_database

is a path to the all_node_db. This is used to determine the shared_memory_prefix.

job_name

This is the job_name for the shared memory that we are clearing; see shared_unique .