In the file /etc/ood/config/clusters.d/brc.yml, if we could implement something like the shell session below, it could get a node from the host list.
OOD doc ref: https://osc.github.io/ood-documentation/latest/installation/resource-manager/linuxhost.html
login:
host: "hpc.brc.berkeley.edu"
ssh_hosts:
- ln000.brc.berkeley.edu
- ln001.brc.berkeley.edu
- ln002.brc.berkeley.edu
- ln003.brc.berkeley.edu
Not sure if below is needed under job section.
bin: "/usr/bin"
conf: "/etc/slurm/slurm.conf"
submit_host: "ln000.brc.berkeley.edu" # ← head node for job submission
ssh_hosts: # ← round-robin login nodes
- ln000.brc.berkeley.edu
- ln001.brc.berkeley.edu
- ln002.brc.berkeley.edu
- ln003.brc.berkeley.edu
strict_host_checking: false # ← needed for round-robin to work
batch_connect:
basic:
script_wrapper: |
module purge
%s
vnc:
script_wrapper: |
module purge
export PATH="/opt/TurboVNC/bin:$PATH"
export WEBSOCKIFY_CMD="/usr/bin/websockify"
%s
In the file /etc/ood/config/clusters.d/brc.yml, if we could implement something like the shell session below, it could get a node from the host list.
OOD doc ref: https://osc.github.io/ood-documentation/latest/installation/resource-manager/linuxhost.html
login:
host: "hpc.brc.berkeley.edu"
ssh_hosts:
- ln000.brc.berkeley.edu
- ln001.brc.berkeley.edu
- ln002.brc.berkeley.edu
- ln003.brc.berkeley.edu
Not sure if below is needed under job section.
batch_connect:
basic:
script_wrapper: |
module purge
%s
vnc:
script_wrapper: |
module purge
export PATH="/opt/TurboVNC/bin:$PATH"
export WEBSOCKIFY_CMD="/usr/bin/websockify"
%s