google/skills
gke-ai-troubleshooting-tpu-mxla-hang
Diagnose GKE Cloud TPU multi-slice training hangs (Megascale `HANG_DETECTED` logs and hang monitored events) using the ML Diagnostics `Megascale XLA (MXLA) Hang Analyzer` (`gcloud alpha mldiagnostics monitored-events`) and 1-minute Cloud Monitoring multi-slice latency metrics (`kubernetes.io/container/multislice/*`). Distinguishes XLA compiler/HLO launch divergence and host data-input stalls from TPU chip, SparseCore, ICI, or network fabric faults. Use when multi-slice TPU training jobs freeze without progressing steps, emit `HANG_DETECTED`, or stall in collective operations. Don't use for gradual step-time throughput drops without hangs (use gke-ai-troubleshooting-tpu-performance-degradation) or pod preemption/eviction restarts (use gke-ai-troubleshooting-jobset-interruption).
Installation
npx skills add https://github.com/google/skills
Similar popular skills
Related neighbors and high-traction skills in the same topics — useful to compare before installing.
Systematic Power BI performance troubleshooting prompt for identifying, diagnosing, and resolvi…
8.9K installsTroubleshoot and resolve common issues with the ClickHouse Node.js client (@clickhouse/client).…
28 installsTroubleshoots failing applications by discovering and analyzing CloudWatch log groups to identi…
31 installsDiagnoses and resolves Amazon S3 Files issues including mount failures, permission errors, sync…
31 installsDiagnoses and resolves Amazon EFS issues including mount failures, NFS timeouts, permission err…
31 installsDiagnoses dbt Cloud/platform job failures by analyzing run logs, querying the Admin API, review…
9 installsAlso in this package
Other skills from google/skills · top by installs.
npx skills add google/skills
More details
Agent compatibility
Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.
Repository health
main