Skip to main content
Open Skills
gke-ai-troubleshooting-tpu-vbar-oomby GoogleOfficialDevelopmentGitHub stars: 21.6k

GKE TPU VBAR OOM Troubleshooting skill: what it does and how to install it

Diagnoses vbar_control_agent crashes and OOMs on GKE TPU v6e nodes by checking Cloud Logging and custom metrics collection.

Summary generated from the skill's documentation.

Install

$ npx skills add google/skills --skill gke-ai-troubleshooting-tpu-vbar-oom

Run it in a terminal. If your agent is already running, start a new session so it picks the skill up.

About this skill

What it does. Guides investigation of TPU v6e failures by defining a time window, checking serial console logs for agent OOMs, and searching for tpu-device-plugin metric checksum errors. It also checks for custom metrics collectors and outlines disabling them or upgrading GKE when a resiliency fix is available.

When to use it. For vbar_control_agent crashes, memory cgroup OOMs, TPU metrics corruption, or custom metrics conflicts on GKE TPU v6e nodes. It is not for general container OOMs or routine GKE node operations.

History

Repo stars

21.6kAbout +3.3k since 30 Jul 2026

Before 1 Oct 2026 the curve is estimated from public event data.

Stars are counted for the whole repository, which holds 130 skills.

Show as a table
Repo stars by day
DateRepo stars
26 Oct 202621,583
25 Oct 202621,560
24 Oct 202621,534
23 Oct 202621,511
22 Oct 202621,483
21 Oct 202621,458
20 Oct 202621,430
19 Oct 202621,411
18 Oct 202621,389
17 Oct 202621,374
16 Oct 202621,351
15 Oct 202621,316
14 Oct 202621,272
13 Oct 202621,235
12 Oct 202621,200
11 Oct 202621,140
10 Oct 202621,101
9 Oct 202621,071
8 Oct 202621,041
7 Oct 202621,006
6 Oct 202620,967
5 Oct 202620,939
4 Oct 202620,924
3 Oct 202620,865
2 Oct 202620,592
1 Oct 202620,557
24 Sept 2026 (estimated)20,498
17 Sept 2026 (estimated)20,452
10 Sept 2026 (estimated)19,878
3 Sept 2026 (estimated)18,833
27 Aug 2026 (estimated)18,776
20 Aug 2026 (estimated)18,765
13 Aug 2026 (estimated)18,765
6 Aug 2026 (estimated)18,294
30 Jul 2026 (estimated)18,271

Installs

4.5k+283 since 19 Oct 2026

Installs via skills.sh

Show as a table
Installs by day
DateInstalls
26 Oct 20264,455
25 Oct 20264,406
24 Oct 20264,384
23 Oct 20264,357
22 Oct 20264,311
21 Oct 20264,263
20 Oct 20264,216
19 Oct 20264,172

Similar skills