Page Comparison

Finalize Integration with Slurm

1. Edit resume_xspot.sh and add a sed command for every pool:

   • Code Block
     user_data=$(cat /opt/slurm/etc/xvm16-_user_data | base64 -w 0)

   • becomes:

   • Code Block
     user_data=$(cat /opt/slurm/etc/exostellar/xvm16-_user_data | sed "s/XSPOT_NODENAME/$host/g" | base64 -w 0)

     N.B.: The cat command works against a user_data script in the ${SLURM_CONF_DIR}/exostellar directory.

2. Edit your slurm.conf and add include statement to pick up xspot.slurm.conf. Replace ${SLURM_CONF_DIR} with the path to the Slurm configuration directory:

   1. Code Block
      include ${SLURM_CONF_DIR}/exostellar/xspot.slurm.conf

3. Verify the xpot.slurm.conf file’s ResumeProgram and SuspendProgram point correctly at ${SLURM_CONF_DIR}/exostellar/resume_xspot.py and ${SLURM_CONF_DIR}/exostellar/suspend_xspot.py.

4. Introducing new nodes into a slurm cluster requires restart of the slurm control deamon:

   1. Code Block
      systemctl restart slurmctld

5. Integration steps are complete and a job submission to the new partition is the last validation:

   1. As a cluster user, navigate to a valid job submission directory and launch a job as normal, but be sure to specify the new partition:

      1. sbatch -p NewPartitionName < job-script.sh