Issues

• SnapshotJob can leave published helper artifacts orphaned after capture failure

  approved
  Feature

  ai-dynamo/snapshot#457 · Filter by author hhzhang16 opened on Oct 5, 2026
  1

• Require maintainer-triggered GPU E2E after successful CPU CI before merging PRs

  Task

  ai-dynamo/snapshot#453 · Filter by author hhzhang16 opened on Oct 5, 2026

• Run CPU checkpoint and restore E2E tests in a fresh cluster on every PR

  Task

  ai-dynamo/snapshot#452 · Filter by author hhzhang16 opened on Oct 5, 2026
  

• Cover CPU SnapshotJob checkpoint and restore in the lifecycle tests

  Task

  ai-dynamo/snapshot#451 · Filter by author hhzhang16 opened on Oct 5, 2026

• Prepare the E2E harness for CPU-only clusters

  Task

  ai-dynamo/snapshot#450 · Filter by author hhzhang16 opened on Oct 5, 2026

• Support checkpointing restored workloads

  Task

  ai-dynamo/snapshot#447 · Filter by author dfeigin-nv opened on Oct 5, 2026

• [DOC]: Add a project roadmap (docs/roadmap.md)

  documentation
  Task

  ai-dynamo/snapshot#415 · Filter by author ozbarshalom opened on Sep 29, 2026
  

• CUDA checkpoint helper hangs with a different libcuda version than the target

  bug
  Bug

  ai-dynamo/snapshot#412 · Filter by author andrewpaprotsky opened on Sep 29, 2026

• SNEP: Application lifecycle hooks for checkpoint/restore

  featurepriority: P0
  Task

  ai-dynamo/snapshot#411 · Filter by author Ronkahn21 opened on Sep 29, 2026
  

• [Design] Retry mechanism for capture failures

  approved
  Task

  ai-dynamo/snapshot#394 · Filter by author danbar2 opened on Sep 23, 2026
  

• build(agent): remove unused placeholder image target

  enhancement
  Task

  ai-dynamo/snapshot#388 · Filter by author julienmancuso opened on Sep 22, 2026

• Add vLLM 0.28 support

  enhancement
  Task

  ai-dynamo/snapshot#384 · Filter by author ozbarshalom opened on Sep 22, 2026
  1

  

• Surface friendly errors instead of propogate inner tools errors

  enhancementgood first issue
  Bug

  ai-dynamo/snapshot#370 · Filter by author ozbarshalom opened on Sep 21, 2026
  1

• Align SnapshotJob launch setup with native multi-GPU jobfile requirements

  bug
  ai-dynamo/snapshot#362 · Filter by author galletas1712 opened on Sep 21, 2026

• Include the Apache-2.0 license in the packaged Helm chart

  approved
  Task

  ai-dynamo/snapshot#358 · Filter by author saturley-hall opened on Sep 17, 2026
  

• Avoid no-op restored condition status writes

  enhancement
  Task

  ai-dynamo/snapshot#353 · Filter by author hutm opened on Sep 17, 2026
  

• Reuse observed Pod metadata on non-DRA restores

  enhancement
  Task

  ai-dynamo/snapshot#352 · Filter by author hutm opened on Sep 17, 2026
  

• Keep Normal Event writes off the restore critical path

  enhancement
  Task

  ai-dynamo/snapshot#351 · Filter by author hutm opened on Sep 17, 2026
  

• Add restore controller phase timing

  enhancement
  Task

  ai-dynamo/snapshot#350 · Filter by author hutm opened on Sep 17, 2026
  

• SNEP: Coordinated multi-Pod snapshot and restore

  approvedenhancement
  Feature

  ai-dynamo/snapshot#324 · Filter by author julienmancuso opened on Sep 16, 2026 · v0.3.0
  

• [Design] Support Fallback to coldstart on restore failures

  approved
  Task

  ai-dynamo/snapshot#322 · Filter by author danbar2 opened on Sep 16, 2026

• Retry mechanism and workload signal on snapshot checkpoint failures

  enhancement
  Feature

  ai-dynamo/snapshot#319 · Filter by author ozbarshalom opened on Sep 16, 2026

• Restore compatibility: refuse a restore onto a different MIG profile

  approved
  Task

  ai-dynamo/snapshot#317 · Filter by author leeZardNav opened on Sep 16, 2026
  

• vLLM restore loops when HTTP health probes are enabled

  bug
  Bug

  ai-dynamo/snapshot#311 · Filter by author dfeigin-nv opened on Sep 15, 2026
  1

  

• Restore reports binfmt_misc instead of the missing NFS mount

  bug
  Bug

  ai-dynamo/snapshot#310 · Filter by author dfeigin-nv opened on Sep 15, 2026
  2