switch_sprintf_node_info.

and times. Limit sizes of strings to avoid having them stray into
other fields.

  use xmalloc/xfree.

  data object in job_info records: slurm_get_select_jobinfo.

Oldl code would result in memory corruption.

NOT_RESPONDING flag from node state.

Move pretty much all BGL-specific logic into that module and the
  associated plugin and make use of an opaque data object for
  maintaining the information.

  one node. NOTE: srun uses hostlist to report task numbers.

 rejected instead of sticking around in a pending state.

removed memory checking bug, and put in fake bgl partition id.  this is a tested (on hal) and working version.

  reported as desired.
Add new function to drain node (for use by select/bluegene node
  monitoring thread).

identification of block allocations.

caught segfaulting error = not correctly freeing memory after using in strtok_r.  Also, cleaned up code so that the defs proceed executable statements

  conn_type, and rotate).
slurmctld to set min_nodes for job based upon geometry (if supplied).
  This avoid trying to schedule jobs when insufficient resources are
  available.