
Gke AI troubleshooting tpu slice mig stuck updating
Diagnoses GKE TPU slice creation stalls caused by unstable Compute Engine MIGs
- Scripts
- None
- Platforms
- Claude Code, Codex, Cursor
- GitHub stars
- 24.4K
- Last commit
- 2 days ago
- License
- Apache-2.0
- Listing updated
- 1 day ago
Similar skills
About Gke AI troubleshooting tpu slice mig stuck updating
Gke AI troubleshooting tpu slice mig stuck updating is an open-source agent skill for cloud and infrastructure. It is available on Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf and Goose.
This skill guides an agent through troubleshooting GKE TPU Dynamic Slice creation and JobSet scheduling delays when the managed instance groups behind a TPU node pool are stuck updating or report isStable=False. It starts by checking the Slice custom resource for a SliceCreationFailed reason, then queries Cloud Logging for errors on the backing instance group manager, and only then inspects MIG stability and version target status.
The workflow narrows down the cause step by step: it looks at individual instance actions and recent instance errors, then checks whether a node pool upgrade or automated repair explains the instability. Capacity and quota errors such as resource pool exhaustion are routed to the matching GKE and TPU troubleshooting sections. The skill is read-only, so it only runs diagnostic commands and leaves any fix for the user to apply.
It expects a GKE cluster with TPU node pools, gcloud and kubectl access, and the Compute, Container and Logging APIs enabled, along with viewer-level IAM roles. Commands are run from a terminal and the skill relies on Google Cloud documentation links for the detailed steps.
When your agent uses it
Diagnose and resolve GKE TPU Dynamic Slice creation and JobSet scheduling failures blocked by underlying Compute Engine managed instance groups (MIGs) stuck in an updating or unstable state (isStable=False). Use when TPU slice creation fails with a SliceCreationFailed status reason, when JobSets remain stuck in pending, or when MIG status reflects isStable=False or versionTarget.isReached=False during TPU slice formation. Don't use for SliceCreationFailed causes unrelated to MIG stability (use gke-ai-troubleshooting-tpu-dynamic-slices-monitoring) or general pod scheduling failures unrelated to TPU slices.
From the skill's SKILL.md. Your agent reads this description to decide when to load the skill.
What it can access
From a static check of its 2 files (17 KB) on GitHub on 17 November 2026. Nothing was run. Read the files before you install a skill: it can tell your agent to run anything its tools allow.
Check another skill before you install it.
Install
Copy the skill folder to ~/.claude/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating to use it in all your projects:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/google/skills.git "$tmp" &&
mkdir -p ~/.claude/skills &&
rm -rf ~/.claude/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
cp -R "$tmp/skills/cloud/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating" ~/.claude/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
rm -rf "$tmp"For one project only, put the folder in .claude/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating inside it instead.
Copy the skill folder to ~/.agents/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating to use it in every repository:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/google/skills.git "$tmp" &&
mkdir -p ~/.agents/skills &&
rm -rf ~/.agents/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
cp -R "$tmp/skills/cloud/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating" ~/.agents/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
rm -rf "$tmp"For one repository only, put the folder in .agents/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating inside it instead.
Copy the skill folder to ~/.cursor/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating to use it in every project:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/google/skills.git "$tmp" &&
mkdir -p ~/.cursor/skills &&
rm -rf ~/.cursor/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
cp -R "$tmp/skills/cloud/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating" ~/.cursor/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
rm -rf "$tmp"For one project only, put the folder in .cursor/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating inside it instead.
Run in a terminal (add --scope workspace to install it for the current workspace only):
gemini skills install https://github.com/google/skills.git --path skills/cloud/gke-ai-troubleshooting-tpu-slice-mig-stuck-updatingOr copy the folder to ~/.gemini/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating yourself.
For GitHub Copilot in VS Code, copy the skill folder to ~/.copilot/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/google/skills.git "$tmp" &&
mkdir -p ~/.copilot/skills &&
rm -rf ~/.copilot/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
cp -R "$tmp/skills/cloud/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating" ~/.copilot/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating &&
rm -rf "$tmp"For one project only, put the folder in .github/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating inside it instead.
Get the folder:
git clone --depth 1 https://github.com/google/skills.gitThen copy skills/skills/cloud/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating into your agent's skills folder as gke-ai-troubleshooting-tpu-slice-mig-stuck-updating. Agents that support SKILL.md read one folder per skill.
What it does
- Slice status inspection
- Checks the Slice custom resource conditions for SliceCreationFailed and identifies the affected partitions and nodes.
- MIG stability checks
- Reads isStable and versionTarget.isReached on the backing managed instance group to confirm whether it is the cause.
- Cloud Logging error queries
- Provides a Logging filter to surface ERROR entries from the instance group manager within a time window.
- Node pool operation review
- Looks at running, cancelled or failed upgrades and automated repairs that could leave the MIG changing.
- Read-only diagnostics
- Runs only read-only commands and gives remediation steps to the user rather than changing the cluster.
- Guided recovery steps
- Points to documented steps for waiting, cancelling, resuming or rolling back node pool upgrades and handling quota errors.
Good for
- Why is my TPU slice stuck in SliceCreationFailed after a node pool upgrade?
- Check whether the managed instance group behind my TPU node pool is stable
- Find out why my JobSet is pending on TPU nodes
- Look up MIG errors such as quota exceeded for my GKE TPU node pool
- Verify that a TPU slice becomes ACTIVE after the MIG finishes updating
Gke AI troubleshooting tpu slice mig stuck updating FAQ
What does this skill need before it runs?
It needs gcloud and kubectl configured for the cluster, an active Cloud Billing account, the container, compute and logging APIs enabled, and the Logs Viewer, Compute Viewer and Kubernetes Engine Viewer roles.
Does it change anything in my cluster?
No. The skill is read-only: it runs diagnostic commands only and never drains, deletes or recreates nodes. Any fix is left for the user to apply.
When should I use this skill instead of the monitoring one?
Use it when the slice failure is tied to MIG stability (isStable=False or versionTarget.isReached=False). For SliceCreationFailed causes unrelated to MIG stability, the skill directs you to the gke-ai-troubleshooting-tpu-dynamic-slices-monitoring skill.
Does it include scripts?
No, the skill has no scripts. It consists of a SKILL.md with instructions and one attribution reference file.
How do I install Gke AI troubleshooting tpu slice mig stuck updating in Claude Code?
Copy the skill folder from GitHub to ~/.claude/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating to use it in all your projects, or to .claude/skills/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating in one project.
Does Gke AI troubleshooting tpu slice mig stuck updating run scripts?
No. It is instructions only, with no scripts. This comes from a static check of the files on GitHub; nothing was run. Read the files before you install it.
Which AI agents can use Gke AI troubleshooting tpu slice mig stuck updating?
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf and Goose. Each reads skills in the SKILL.md format from its own folder; the install steps on this page list them.
What are the best alternatives to Gke AI troubleshooting tpu slice mig stuck updating?
The closest alternatives to Gke AI troubleshooting tpu slice mig stuck updating are Gke AI troubleshooting tpu partition overlap, Gke AI troubleshooting tpu dynamic slices monitoring, Gke AI troubleshooting tpu vm repair stuck, Gke AI troubleshooting tpu reservation stockout and Gke AI troubleshooting tpu mxla hang.
Where this comes from
- Description: from the skill's own SKILL.md.
More from google/skills

Route platform choices to Google's own skills by fetching and filtering its skill catalog
- Open source
- 24.4K GitHub stars
- Cloud and infrastructure

Guardrails for gcloud commands: validate syntax with help, scope projects, dry-run first
- Open source
- 24.4K GitHub stars
- Cloud and infrastructure

Authenticate to Google Cloud: ADC, impersonation, service accounts and IAM roles
- Open source
- 24.4K GitHub stars
- Cloud and infrastructure

First-time Google Cloud setup: sign in, pick a project, link billing, with consent gates
- Open source
- 24.4K GitHub stars
- Cloud and infrastructure

Look up Google developer docs through the Developer Knowledge MCP server or REST fallback
- Open source
- 24.4K GitHub stars
- Developer tools

Add, remove or clear audience members with the Google Data Manager API
- Open source
- 24.4K GitHub stars
- Marketing and SEO



