Skip to content

sparkrun status

Prefer a saved default cluster or --cluster NAME for host selection, including single-node use. See Recommended Cluster Setups.

Terminal window
sparkrun status [options]

Show sparkrun workloads reported by enabled executors on cluster hosts. Displays containers grouped by job with their role, host, status, image, and ready-to-use logs and stop commands.

OptionDescription
--hosts / -HComma-separated host list
--hosts-fileFile with hosts (one per line, # comments)
--clusterUse a saved cluster by name
--jsonEmit machine-readable JSON instead of the table
--dry-run / -nShow what would be done

If no options are given, uses the default cluster.

  • Running jobs — workloads grouped by workload ID, with role (head/worker/solo), host, status, and image
  • IP mapping — when InfiniBand was detected, shows the complementary IP (management or IB) alongside each host so you can identify both addresses
  • Idle hosts — hosts with no sparkrun workloads
  • Pending operations — active downloads or image distributions that haven’t launched containers yet, with elapsed time. These will consume VRAM once launched
  • Copy-paste commands — ready-to-use sparkrun logs and sparkrun stop commands for each job
Terminal window
# Check default cluster
sparkrun status
# Check a specific cluster
sparkrun status --cluster mylab
# Check a saved subgroup
sparkrun status --cluster pair-a