-
Notifications
You must be signed in to change notification settings - Fork 48
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
- Status: Open.#967 In ROCm/spur;
- Status: Open.#965 In ROCm/spur;
Single-task-per-node srun with --gpus-per-node=8 leaves /dev/dri empty (0 GPUs visible)
area: job-lifecycleJob submission, states, env varsJob submission, states, env varsarea: resourcesGPU, CPU, memory, topology, cgroupsGPU, CPU, memory, topology, cgroupspriority: highImportant for compatibilityImportant for compatibilityslurm-compatSlurm compatibility gapSlurm compatibility gaptype: bugSomething isn't workingSomething isn't workingStatus: Open.#962 In ROCm/spur;Short-lived srun --pty command intermittently fails with a PID-report timeout
type: bugSomething isn't workingSomething isn't workingtype: testTest improvements or additionsTest improvements or additionsStatus: Open.#952 In ROCm/spur;Managed k0s nodes cannot share their GPUs between Spur jobs and pods
area: resourcesGPU, CPU, memory, topology, cgroupsGPU, CPU, memory, topology, cgroupsarea: schedulingPriority, backfill, fairshare, preemptionPriority, backfill, fairshare, preemptionenhancementNew feature or requestNew feature or requestStatus: Open.#923 In ROCm/spur;- Status: Open.#922 In ROCm/spur;
- Status: Open.#920 In ROCm/spur;
- Status: Open.#916 In ROCm/spur;
- Status: Open.#892 In ROCm/spur;
- Status: Open.#891 In ROCm/spur;
- Status: Open.#888 In ROCm/spur;
srun --jobid --overlapattaches via PTY and ignores--mpi=pmixarea: clisbatch, squeue, sinfo, sacct, scontrolsbatch, squeue, sinfo, sacct, scontrolarea: networkingWireGuard mesh, MPI, PMIxWireGuard mesh, MPI, PMIxpriority: highImportant for compatibilityImportant for compatibilityslurm-compatSlurm compatibility gapSlurm compatibility gaptype: bugSomething isn't workingSomething isn't workingStatus: Open.#876 In ROCm/spur;