This update fixes a double-whammy kind of bug. Basically, the L1H temp file (contains enforced displacement data) didn't survive into the second step of a buckling problem. That should cause a fatal error to be issued, but guess what? The subroutine attempting the open doesn't set OUNT, the variable containing the unit numbers for message output files (ERR and F06), so it'd contain uninitialised integers. The result? 50% chance it was negative, causing a runtime error. It being positive caused a memory error and a likely crash.
This update fixes a double-whammy kind of bug. Basically, the L1H temp file (contains enforced displacement data) didn't survive into the second step of a buckling problem. That should cause a fatal error to be issued, but guess what? The subroutine attempting the open doesn't set
OUNT, the variable containing the unit numbers for message output files (ERR and F06), so it'd contain uninitialised integers. The result? 50% chance it was negative, causing a runtime error. It being positive caused a memory error and a likely crash.Check the commit-level data for details.