diff --git a/doc/html/faq.shtml b/doc/html/faq.shtml index b7a9d4005a85a784b170e79599b451f712fe76e1..501f1abc074001dc0eeb37a512173cfb6cf4ac01 100644 --- a/doc/html/faq.shtml +++ b/doc/html/faq.shtml @@ -1739,7 +1739,7 @@ It has integration with Slurm as well as Torque resource managers.</p> <p><a name="add_nodes"><b>47. What process should I follow to add nodes to Slurm?</b></a></br> The slurmctld daemon has a multitude of bitmaps to track state of nodes and cores in the system. Adding nodes to a running system would require the slurmctld daemon -rebuild all of those bitmaps, which the developers feel would be safer to do by +to rebuild all of those bitmaps, which the developers feel would be safer to do by restarting the daemon. Communications from the slurmd daemons on the compute nodes to the slurmctld daemon include a configuration file checksum, so you probably also want to maintain a common slurm.conf file on all nodes. The @@ -1753,6 +1753,10 @@ following procedure is recommended: "scontrol reconfig", no need to restart the daemons)</li> </ol> +NOTE: Jobs submitted with srun, and that are waiting for an allocation, +prior to new nodes being added to the slurm.conf can fail if the job is +allocated one of the new nodes. + <p><a name="licenses"><b>48. Can Slurm be configured to manage licenses?</b></a></br> Slurm is not currently integrated with FlexLM, but it does provide for the allocation of global resources called licenses. Use the Licenses configuration diff --git a/doc/html/mpi_guide.shtml b/doc/html/mpi_guide.shtml index 546edcf744c571e35d32396c1e410a96454db67d..581eacd479b5a25658f29364d1c92a9966248ada 100644 --- a/doc/html/mpi_guide.shtml +++ b/doc/html/mpi_guide.shtml @@ -14,7 +14,7 @@ mpirun launches tasks using Slurm's infrastructure (LAM/MPI and HP-MPI).</li> <li>Slurm creates a resource allocation for the job and then mpirun launches tasks using some mechanism other than SLURM, such as SSH or RSH (BlueGene MPI and some MPICH1 modes). -These tasks initiated outside of Slurm's monitoring +These tasks are initiated outside of Slurm's monitoring or control. Slurm's epilog should be configured to purge these tasks when the job's allocation is relinquished. </li> </ol>