[CI] lmstudio-rocm-llms / lmstudio-load-model-linux failed on halo (linux)

[github-actions]

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

Failure scope

• Playbook: lmstudio-rocm-llms
• Test id: lmstudio-load-model-linux
• Device: halo
• Operating system: linux
• Runner labels: self-hosted, Linux, halo
• Runner name: xsj-aimlab-halo-03
• Commit: 36cd64215e5662a8880a9ae25b85e2f66edec256
• Workflow run: https://github.com/amd/playbooks/actions/runs/28559316323

Hardware / OS to use to reproduce

Run the failing test on a machine that matches the runner labels above (OS = linux, device = halo). 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=lmstudio-rocm-llms

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 lmstudio-rocm-llms --platform linux --device halo

The runner extracts test blocks from playbooks/*/lmstudio-rocm-llms/README.md (the failing block starts around line 359).

Failing test (verbatim from the README)

• Timeout: 1200s

lms unload --all || true
lms ps
ID="gpt-oss-120b-${GITHUB_RUN_ID}"
echo "$ID" > /tmp/lmstudio_model_id.txt
lms load gpt-oss-120b --context-length 32768 --gpu max --identifier "$ID"
lms ps # Verify model is really loaded
lms chat "$ID" -p "Reply with exactly: OK"

Result

• Exit code: 1

stderr (last lines)

No models to unload.
No models are currently loaded.

To load a model, run:

    lms load <model path>
Error: Error loading model. 


 (!) SUGGESTION 


   (X) CAUSE  

(Exit code: null). Please check settings and try loading the model again.
Error: Model "gpt-oss-120b-28559316323" not found, load with:
       lms load gpt-oss-120b-28559316323

stdout (last lines)

Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴�[?25l
Loading gpt-oss-120b ⠦�[?25l
Loading gpt-oss-120b ⠧�[?25l
Loading gpt-oss-120b ⠇�[?25l
Loading gpt-oss-120b ⠏�[?25l
Loading gpt-oss-120b ⠋�[?25l
Loading gpt-oss-120b ⠙�[?25l
Loading gpt-oss-120b ⠹�[?25l
Loading gpt-oss-120b ⠸�[?25l
Loading gpt-oss-120b ⠼�[?25l
Loading gpt-oss-120b ⠴
�[K�[?25h
IDENTIFIER                  MODEL           STATUS    SIZE        CONTEXT    DEVICE    TTL
gpt-oss-120b-28559316323    gpt-oss-120b    IDLE      63.39 GB    32768      Local        


�[K�[?25h

stdout was truncated; see the workflow run artifacts for the full log.

---

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.

[github-actions]

Test failed again on a subsequent CI run.

• Workflow run: https://github.com/amd/playbooks/actions/runs/28690839531
• Commit: 36cd64215e5662a8880a9ae25b85e2f66edec256
• Runner: xsj-aimlab-halo-03 (self-hosted, Linux, halo)

[github-actions]

Test failed again on a subsequent CI run.

• Workflow run: https://github.com/amd/playbooks/actions/runs/28725753788
• Commit: 36cd64215e5662a8880a9ae25b85e2f66edec256
• Runner: xsj-aimlab-halo-03 (self-hosted, Linux, halo)

[github-actions]

Test failed again on a subsequent CI run.

• Workflow run: https://github.com/amd/playbooks/actions/runs/28762169190
• Commit: 36cd64215e5662a8880a9ae25b85e2f66edec256
• Runner: xsj-aimlab-halo-03 (self-hosted, Linux, halo)

[github-actions]

Test failed again on a subsequent CI run.

• Workflow run: https://github.com/amd/playbooks/actions/runs/28987705212
• Commit: 52dde5e996c09dd126a0e10337e41bc5b91d1945
• Runner: xsj-aimlab-halo-02 (self-hosted, Linux, halo)

[github-actions]

Test failed again on a subsequent CI run.

• Workflow run: https://github.com/amd/playbooks/actions/runs/29175365761
• Commit: ca4ed8f5015b90479e1c103be4d919dd01a10169
• Runner: xsj-aimlab-halo-03 (self-hosted, Linux, halo)

[github-actions]

Test failed again on a subsequent CI run.

• Workflow run: https://github.com/amd/playbooks/actions/runs/29319302529
• Commit: 64f23e95a0e186300e7fc7f2aebb063c6b171eec
• Runner: xsj-aimlab-halo-03 (self-hosted, Linux, halo)

[lucbruni-amd]

Root cause: this occurred while the halo Linux runners had LM Studio's CPU (avx2) runtime selected, which cannot reliably load the 63GB gpt-oss-120b model. The recorded failure is a load crash (Error loading model, exit code null), not a timeout. The halo Linux runners were switched to the Vulkan runtime; reproduced 3/3 clean loads (~60s each, disk-bound) with GPU inference ~2.3s. PR #549 pins a GPU runtime in CI so this cannot silently regress. Closing as root-caused and fixed; reopen if it recurs.