How-To

Articles

• How-To Capture NVIDIA Bug Report via Command Center
• How To Enable GPU Direct Storage (GDS) on Crusoe GPU Instances
• How To Download Large Models from AWS S3 to Local NVMe Using rclone
• How-To Setup GB200 NVL72 Rack on CMK Cluster and Run NCCL Performance Validation
• How-To Validate Infiniband Performance with NCCL All Reduce Test
• How-To Setup Serial-Console to Access Your VM
• How-To Enable Node Pool Taints on CMK
• How-To Import an Existing Partition as crusoe_transport_partition
• How-To Collect Software RAID Diagnostics
• How-To Collect NVMe Drive Diagnostics
• How-To Migrate crusoe_ib_partition to crusoe_transport_partition in Terraform using moved {} block
• How-To Enable NVIDIA MPS GPU Sharing on Crusoe Managed Kubernetes
• How-To Enable GPU Time-Slicing on Crusoe Managed Kubernetes
• How-To Recover a CMK Node When a Full Boot Disk Blocks Access
• How-To Update Image Pull Timeouts for Large Container Images on CMK
• How-To Map Guest NVMe Devices to PCIe BDF Topology
• How-To Configure NVIDIA Device Plugin to Ignore Specific XID Errors Using Helm
• How-To Safely Upgrade the VAST NFS Driver on a Crusoe VM
• How-To Fix Shared Disks Mounting Read-Only Despite the rw Option
• How-To Safely Attach a Boot Disk from a Same-Image VM Without Boot Conflicts
• How-To Restrict Internet Access From VMs and Find Ports That Need to be Open
• How-To Setup RAID0 Storage on CMK with Ephemeral Storage for Containerd
• How-To: Add SSH Keys to CMK Nodes After Node Pool Creation
• How-To Enable NCCL Debug Logging for Collective Timeout Failures
• How-To Recover a Slurm Node Drained Due to Full Root Disk
• How-To Diagnose nvidia.com/hostdev: 0 on CMK Nodes
• How-To Detect Silent Data Corruption in Distributed Training Jobs
• How To Fix Missing GPU Metrics (DCGM) on CMK Nodes
• How-To: Add or Update Lifecycle Scripts on an Existing Crusoe VM
• How-To: Validate Infiniband Performance with RCCL All Reduce Test