read_restart with large timesteps


The read_restart command with a '*' pattern doesn't work for 64-bit timesteps. For instance, with a file "restart.2500000000":

read_restart restart.*

in the input script gives:

ERROR on proc 0: Found no restart file matching pattern

but this:

read_restart restart.2500000000

works fine with no error.

I'll check it out. Did you really run for
2.5 billion timesteps, or did you just set
up this test to look awesome ?


Fixed it - see the 29Jul11 patch.
This slipped by when we upgraded the timestep
counter to be a 64-bit integer.