What's Changed
- Slurm Update (24.05.1)
- Slurm nodes are dynamic instead of pre-defined (+ custom hostnames)
- Multi MT FSS deployment (Using DNS Round Robin)
- New monitoring solution (Prometheus, Node exporter, Can be run on a separate instance)
- Fix to slurm Healthchecks
- New shapes added, MI300X, L40S, A100 VMs,....
- OpenLDAP fix for OL8
- Meshpinger added (Used to validate RDMA connectivity between all RDMA NICs on all hosts)
- NVME's (Switch from LVM to MDADM)
- New OL8 images (OCA hold)
- Bug Fixes