[CI] vscode-qwen3-coder / lmstudio-load-qwen3-coder-linux failed on krk (linux)

[github-actions]

This issue was opened automatically by the Test Playbooks workflow after the test lmstudio-load-qwen3-coder-linux failed on the main branch.

Failure scope

• Playbook: vscode-qwen3-coder
• Test id: lmstudio-load-qwen3-coder-linux
• Device: krk
• Operating system: linux
• Runner labels: self-hosted, Linux, krk
• Runner name: xsj-aimlab-krk-04
• Commit: 60fb4cac864b0a4e8451b80518797f38b0c61ba0
• Workflow run: https://github.com/amd/playbooks/actions/runs/30080636481

Hardware / OS to use to reproduce

Run the failing test on a machine that matches the runner labels above (OS = linux, device = krk). The repo's self-hosted runners already advertise these labels; if you reproduce locally, use the same OS family and the same AMD device class.

How to dispatch the same test from CI

Re-run only the failing playbook on the same matrix entry by triggering the workflow with the playbook id:

gh workflow run test-playbooks.yml --repo amd/playbooks -f playbook_id=vscode-qwen3-coder

The workflow's matrix narrows down to this (device, platform) combination automatically based on the playbook's tested_platforms.

How to run just this test locally

python .github/scripts/run_playbook_tests.py --playbook vscode-qwen3-coder --platform linux --device krk

The runner extracts test blocks from playbooks/*/vscode-qwen3-coder/README.md (the failing block starts around line 370).

Failing test (verbatim from the README)

• Timeout: 1200s

lms unload --all || true
lms ps
ID="qwen3coder-32k-${GITHUB_RUN_ID}"
echo "$ID" > /tmp/lmstudio_model_id.txt
# retry once: large-model loads can transiently fail under memory pressure
lms load qwen3-coder-30b --context-length 32768 --gpu max --identifier "$ID" -y || { lms unload --all; sleep 5; lms load qwen3-coder-30b --context-length 32768 --gpu max --identifier "$ID" -y; }
lms ps # Verify model is really loaded
lms chat "$ID" -p "Reply with exactly: OK"

Result

• Exit code: -1
• Runner error: Test timed out after 1200 seconds

---

This issue is opened and deduplicated by .github/scripts/create_failure_issues.py. Close it once the failure is fixed; subsequent failures with the same scope will reopen a fresh issue.