Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Select an option

  • Save hexfusion/58df60808cb34b02e2ed41a5e391fd61 to your computer and use it in GitHub Desktop.

Select an option

Save hexfusion/58df60808cb34b02e2ed41a5e391fd61 to your computer and use it in GitHub Desktop.
# CKS nightly capacity-related failures
# Collected: 2026-05-26
# Repo: llm-d/llm-d
#
# How this was generated:
# ./scripts/llm-d-ci-capacity-failures.sh CKS 200 scripts/cks-capacity-failures-2026-05-26.txt
#
# Equivalent inline command:
# gh run list --repo llm-d/llm-d --limit 200 --status failure \
# --json databaseId,workflowName,createdAt \
# --jq '.[] | select((.workflowName | contains("CKS")) and (.workflowName | contains("Nightly"))) | "\(.databaseId)\t\(.createdAt)\t\(.workflowName)"' \
# | while IFS=$'\t' read -r id ts name; do
# if gh run view "$id" --repo llm-d/llm-d --log-failed 2>/dev/null \
# | grep -qE '##\[error\]Pods in .* did not become ready within|##\[error\]Insufficient GPUs|FailedScheduling.*Insufficient nvidia\.com/gpu'; then
# echo "$ts https://github.com/llm-d/llm-d/actions/runs/$id $name"
# fi
# done
#
# Capacity signals matched (real GitHub Actions error annotations only,
# not bash-echo script bodies):
# - ##[error]Pods in <ns> did not become ready within <duration>
# - ##[error]Insufficient GPUs: ...
# - FailedScheduling event paired with Insufficient nvidia.com/gpu
#
# Result: 16 capacity failures in last 200 CKS nightly runs.
2026-05-26T10:21:56Z https://github.com/llm-d/llm-d/actions/runs/26446592490 Nightly - Precise Prefix Cache E2E (CKS)
2026-05-26T08:11:33Z https://github.com/llm-d/llm-d/actions/runs/26440503176 Nightly - Predicted Latency E2E (CKS)
2026-05-25T15:55:12Z https://github.com/llm-d/llm-d/actions/runs/26409007223 Nightly - Precise Prefix Cache E2E (CKS)
2026-05-25T08:28:51Z https://github.com/llm-d/llm-d/actions/runs/26391191536 Nightly - Predicted Latency E2E (CKS)
2026-05-24T07:54:58Z https://github.com/llm-d/llm-d/actions/runs/26355682668 Nightly - Predicted Latency E2E (CKS)
2026-05-23T07:46:53Z https://github.com/llm-d/llm-d/actions/runs/26327306158 Nightly - Predicted Latency E2E (CKS)
2026-05-20T15:11:34Z https://github.com/llm-d/llm-d/actions/runs/26171642277 Nightly - Precise Prefix Cache E2E (CKS)
2026-05-20T15:11:19Z https://github.com/llm-d/llm-d/actions/runs/26171628516 Nightly - optimized baseline E2E (CKS)
2026-05-20T08:37:07Z https://github.com/llm-d/llm-d/actions/runs/26151265275 Nightly - WVA E2E (CKS)
2026-05-19T07:17:24Z https://github.com/llm-d/llm-d/actions/runs/26082364917 Nightly - optimized baseline E2E (CKS)
2026-05-18T08:22:51Z https://github.com/llm-d/llm-d/actions/runs/26021987902 Nightly - Wide EP LWS E2E (CKS)
2026-05-18T08:19:07Z https://github.com/llm-d/llm-d/actions/runs/26021820841 Nightly - Predicted Latency E2E (CKS)
2026-05-18T07:28:24Z https://github.com/llm-d/llm-d/actions/runs/26019629028 Nightly - optimized baseline E2E (CKS)
2026-05-17T07:50:04Z https://github.com/llm-d/llm-d/actions/runs/25985082951 Nightly - Wide EP LWS E2E (CKS)
2026-05-17T07:47:15Z https://github.com/llm-d/llm-d/actions/runs/25985028685 Nightly - Predicted Latency E2E (CKS)
2026-05-16T07:37:54Z https://github.com/llm-d/llm-d/actions/runs/25956345187 Nightly - Wide EP LWS E2E (CKS)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment