[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-mcore-create-issue":3,"mdc--t4d8ef-key":36,"related-repo-nvidia-mcore-create-issue":1674,"related-org-nvidia-mcore-create-issue":1759},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":22,"repoUrl":23,"updatedAt":24,"license":25,"forks":26,"topics":27,"repo":31,"sourceUrl":34,"mdContent":35},"mcore-create-issue","investigate CI failures and create issues","Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19],{"name":13,"slug":14,"type":15},"GitHub","github","tag",{"name":17,"slug":18,"type":15},"Triage","triage",{"name":20,"slug":21,"type":15},"Debugging","debugging",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-14T05:25:57.442089","Apache-2.0",4230,[28,29,30],"large-language-models","model-para","transformers",{"repoUrl":23,"stars":22,"forks":26,"topics":32,"description":33},[28,29,30],"Ongoing research training transformer models at scale","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM\u002Ftree\u002FHEAD\u002Fskills\u002Fmcore-create-issue","---\nname: mcore-create-issue\ndescription: Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.\nlicense: Apache-2.0\nwhen_to_use: User shares a GitHub Actions URL and wants to file a bug report; 'create an issue for this failure', 'file a bug for this CI run', 'triage this GitHub Actions failure'.\nuser_invocable: true\nargument: \"GitHub Actions run or job URL\"\nmetadata:\n  author: Oliver Koenig \u003Cokoenig@nvidia.com>\n---\n\n# Triage CI Failure into a GitHub Issue\n\nInvestigate a failing GitHub Actions job, extract the root cause, and file a\nwell-structured bug issue against `NVIDIA\u002FMegatron-LM`.\n\n## Workflow\n\n### 1. Parse the URL\n\nThe argument is a GitHub Actions URL. It will be one of:\n\n- **Job URL**: `https:\u002F\u002Fgithub.com\u002F\u003Cowner>\u002F\u003Crepo>\u002Factions\u002Fruns\u002F\u003Crun_id>\u002Fjob\u002F\u003Cjob_id>`\n- **Run URL**: `https:\u002F\u002Fgithub.com\u002F\u003Cowner>\u002F\u003Crepo>\u002Factions\u002Fruns\u002F\u003Crun_id>`\n\nExtract `run_id` and, if present, `job_id`.\n\n### 2. Identify failed jobs\n\n- If a `job_id` was provided, use that job directly.\n- If only a `run_id` was provided, list all failed jobs in the run:\n\n  ```bash\n  gh run view \u003Crun_id> --repo NVIDIA\u002FMegatron-LM --json jobs \\\n    --jq '[.jobs[] | select(.conclusion == \"failure\") | {id: .databaseId, name: .name, url: .url}]'\n  ```\n\n  If multiple jobs failed, ask the user which one to triage, or triage all of them if they say so.\n\n### 3. Fetch the failure logs\n\nFor each failed job, retrieve the logs and narrow them down to the failure:\n\n```bash\n# Pull the raw log and keep only error-bearing lines\ngh api repos\u002FNVIDIA\u002FMegatron-LM\u002Factions\u002Fjobs\u002F\u003Cjob_id>\u002Flogs 2>&1 \\\n  | grep -E \"(FAILED|ERROR|\\bError\\b|assert|Traceback|Exception|##\\[error\\])\" \\\n  | head -200\n```\n\nAlso capture the full job name:\n\n```bash\ngh run view --job \u003Cjob_id> --repo NVIDIA\u002FMegatron-LM --json name --jq .name\n```\n\nIf the grep output is sparse, download the full logs and look for the pytest\n`FAILURES` section or the last non-zero exit signal.\n\n### 4. Resolve the triggering PR and test author\n\n**Triggering PR**: the run's head branch follows the pattern `pull-request\u002F\u003Cnumber>`.\nExtract it and resolve the PR:\n\n```bash\ngh run view \u003Crun_id> --repo NVIDIA\u002FMegatron-LM --json headBranch --jq .headBranch\n# → e.g. \"pull-request\u002F4332\"\n# Extract PR number and fetch metadata:\ngh pr view \u003Cpr_number> --repo NVIDIA\u002FMegatron-LM --json number,title,url \\\n  --jq '{number: .number, title: .title, url: .url}'\n```\n\n**Test file author**: find the GitHub login of whoever last touched the failing\ntest file. The file may not exist on `main` — first determine the PR's base\nbranch, then search from there:\n\n```bash\n# 1. Get the PR's base branch (e.g. \"main\", \"dev\", \"release\u002FX.Y\")\ngh pr view \u003Cpr_number> --repo NVIDIA\u002FMegatron-LM --json baseRefName --jq .baseRefName\n\n# 2. Search commits on that base branch\ngh api \"repos\u002FNVIDIA\u002FMegatron-LM\u002Fcommits?path=\u003Ctest-file-path>&sha=\u003Cbase-branch>&per_page=1\" \\\n  --jq '.[0] | {login: .author.login, name: .commit.author.name, sha: .sha}'\n```\n\nIf the result is empty (file was introduced by the PR itself), query the PR's\ncommits instead:\n\n```bash\ngh api \"repos\u002FNVIDIA\u002FMegatron-LM\u002Fpulls\u002F\u003Cpr_number>\u002Fcommits\" \\\n  --jq '[.[] | select(.files? \u002F\u002F [] | any(.filename == \"\u003Ctest-file-path>\"))] | .[0].author.login'\n```\n\nAs a last resort, list the PR commits and pick the author of the commit whose\nmessage most closely relates to the failing test file.\n\n### 5. Extract the root cause\n\nFrom the logs, identify:\n\n- **Failed test(s)**: lines matching `FAILED tests\u002F...::...` give the exact pytest node IDs.\n- **Error message**: the assertion failure, exception type, or first meaningful\n  traceback frame — keep it under ~30 lines.\n- **Job name**: the GitHub Actions job name (e.g. `tests\u002Funit_tests\u002Ftransformer\u002Fmoe\u002F**\u002F*.py - latest`).\n- **Run \u002F job URLs** and **PR URL**: for linking in the issue.\n\n### 6. Check for duplicate issues\n\nSearch for open issues that already cover the same test:\n\n```bash\ngh issue list --repo NVIDIA\u002FMegatron-LM \\\n  --state open \\\n  --search \"\u003Cfailed-test-filename>\" \\\n  --json number,title,url \\\n  --limit 10\n```\n\n- If a matching open issue exists, **do not create a new one**. Report the\n  existing issue to the user and stop.\n- If no match is found, proceed to file a new issue.\n\n### 7. Create the issue\n\nPass `--assignee \u003Ctest-author-login>` to assign the issue to the test file's\nauthor. Include the triggering PR URL in the issue body.\n\n```bash\ngh issue create \\\n  --repo NVIDIA\u002FMegatron-LM \\\n  --title \"🐛 CI failure: \u003Cfailed-test-node-id>\" \\\n  --label \"bug\" \\\n  --assignee \"\u003Ctest-author-login>\" \\\n  --body \"...\"\n```\n\nUse the bug-report template body structure:\n\n```markdown\n**Describe the bug**\n\nCI test `\u003Cfailed-test-node-id>` failed in job [`\u003Cjob-name>`](\u003Cjob-url>).\nTag @NVIDIA\u002Fmcore-oncall to get oncall's attention to this issue.\n\n**Failing run**\n\n| Field | Value |\n|-------|-------|\n| PR    | [#\u003Cpr_number>: \u003Cpr_title>](\u003Cpr_url>) |\n| Run   | [\u003Crun_id>](\u003Crun_url>) |\n| Job   | [\u003Cjob_name>](\u003Cjob_url>) |\n\n**Error**\n\n```\n\u003Ccore error message \u002F traceback — 30 lines max>\n```\n\n**Steps\u002FCode to reproduce bug**\n\nRe-run the failing CI job linked above, or locally inside the dev container:\n\n```bash\npytest \u003Cfailed-test-node-id>\n```\n\n**Additional context**\n\nTriaged automatically via `\u002Ftriage-issue`.\n```\n\nIf multiple tests failed in the same job, list each one as a separate bullet\nunder \"Describe the bug\" and include the combined error snippets. Assign the\nissue to the author of whichever test file appears first in the failure list.\n\n### 8. Report back to the user\n\nPrint the URL of the newly created issue (or the duplicate, if found) so the\nuser can review or share it.\n\n## Important guidelines\n\n- Never create an issue if a duplicate already exists — link the existing one instead.\n- Always include the triggering PR link in the issue body.\n- Always assign the issue to the test file's most recent author. If the author\n  lookup fails (e.g. the commit was made by a bot or the login is unavailable),\n  skip `--assignee` and note it in the \"Additional context\" section.\n- Keep the error snippet concise (≤30 lines). Truncate long tracebacks and note that the full log is available via the job URL.\n- Do not guess the root cause — quote the actual log output verbatim.\n- If the job is still in progress or the logs are unavailable, say so and ask the user to retry once the run completes.\n",{"data":37,"body":43},{"name":4,"description":6,"license":25,"when_to_use":38,"user_invocable":39,"argument":40,"metadata":41},"User shares a GitHub Actions URL and wants to file a bug report; 'create an issue for this failure', 'file a bug for this CI run', 'triage this GitHub Actions failure'.",true,"GitHub Actions run or job URL",{"author":42},"Oliver Koenig \u003Cokoenig@nvidia.com>",{"type":44,"children":45},"root",[46,55,70,77,84,89,126,146,152,290,296,301,422,427,497,510,516,534,690,708,845,850,905,910,916,921,987,993,998,1106,1126,1132,1145,1285,1290,1624,1629,1639,1647,1659,1668],{"type":47,"tag":48,"props":49,"children":51},"element","h1",{"id":50},"triage-ci-failure-into-a-github-issue",[52],{"type":53,"value":54},"text","Triage CI Failure into a GitHub Issue",{"type":47,"tag":56,"props":57,"children":58},"p",{},[59,61,68],{"type":53,"value":60},"Investigate a failing GitHub Actions job, extract the root cause, and file a\nwell-structured bug issue against ",{"type":47,"tag":62,"props":63,"children":65},"code",{"className":64},[],[66],{"type":53,"value":67},"NVIDIA\u002FMegatron-LM",{"type":53,"value":69},".",{"type":47,"tag":71,"props":72,"children":74},"h2",{"id":73},"workflow",[75],{"type":53,"value":76},"Workflow",{"type":47,"tag":78,"props":79,"children":81},"h3",{"id":80},"_1-parse-the-url",[82],{"type":53,"value":83},"1. Parse the URL",{"type":47,"tag":56,"props":85,"children":86},{},[87],{"type":53,"value":88},"The argument is a GitHub Actions URL. It will be one of:",{"type":47,"tag":90,"props":91,"children":92},"ul",{},[93,111],{"type":47,"tag":94,"props":95,"children":96},"li",{},[97,103,105],{"type":47,"tag":98,"props":99,"children":100},"strong",{},[101],{"type":53,"value":102},"Job URL",{"type":53,"value":104},": ",{"type":47,"tag":62,"props":106,"children":108},{"className":107},[],[109],{"type":53,"value":110},"https:\u002F\u002Fgithub.com\u002F\u003Cowner>\u002F\u003Crepo>\u002Factions\u002Fruns\u002F\u003Crun_id>\u002Fjob\u002F\u003Cjob_id>",{"type":47,"tag":94,"props":112,"children":113},{},[114,119,120],{"type":47,"tag":98,"props":115,"children":116},{},[117],{"type":53,"value":118},"Run URL",{"type":53,"value":104},{"type":47,"tag":62,"props":121,"children":123},{"className":122},[],[124],{"type":53,"value":125},"https:\u002F\u002Fgithub.com\u002F\u003Cowner>\u002F\u003Crepo>\u002Factions\u002Fruns\u002F\u003Crun_id>",{"type":47,"tag":56,"props":127,"children":128},{},[129,131,137,139,145],{"type":53,"value":130},"Extract ",{"type":47,"tag":62,"props":132,"children":134},{"className":133},[],[135],{"type":53,"value":136},"run_id",{"type":53,"value":138}," and, if present, ",{"type":47,"tag":62,"props":140,"children":142},{"className":141},[],[143],{"type":53,"value":144},"job_id",{"type":53,"value":69},{"type":47,"tag":78,"props":147,"children":149},{"id":148},"_2-identify-failed-jobs",[150],{"type":53,"value":151},"2. Identify failed jobs",{"type":47,"tag":90,"props":153,"children":154},{},[155,167],{"type":47,"tag":94,"props":156,"children":157},{},[158,160,165],{"type":53,"value":159},"If a ",{"type":47,"tag":62,"props":161,"children":163},{"className":162},[],[164],{"type":53,"value":144},{"type":53,"value":166}," was provided, use that job directly.",{"type":47,"tag":94,"props":168,"children":169},{},[170,172,177,179,284,288],{"type":53,"value":171},"If only a ",{"type":47,"tag":62,"props":173,"children":175},{"className":174},[],[176],{"type":53,"value":136},{"type":53,"value":178}," was provided, list all failed jobs in the run:",{"type":47,"tag":180,"props":181,"children":186},"pre",{"className":182,"code":183,"language":184,"meta":185,"style":185},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","gh run view \u003Crun_id> --repo NVIDIA\u002FMegatron-LM --json jobs \\\n  --jq '[.jobs[] | select(.conclusion == \"failure\") | {id: .databaseId, name: .name, url: .url}]'\n","bash","",[187],{"type":47,"tag":62,"props":188,"children":189},{"__ignoreMap":185},[190,260],{"type":47,"tag":191,"props":192,"children":195},"span",{"class":193,"line":194},"line",1,[196,202,208,213,219,224,230,235,240,245,250,255],{"type":47,"tag":191,"props":197,"children":199},{"style":198},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[200],{"type":53,"value":201},"gh",{"type":47,"tag":191,"props":203,"children":205},{"style":204},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[206],{"type":53,"value":207}," run",{"type":47,"tag":191,"props":209,"children":210},{"style":204},[211],{"type":53,"value":212}," view",{"type":47,"tag":191,"props":214,"children":216},{"style":215},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[217],{"type":53,"value":218}," \u003C",{"type":47,"tag":191,"props":220,"children":221},{"style":204},[222],{"type":53,"value":223},"run_i",{"type":47,"tag":191,"props":225,"children":227},{"style":226},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[228],{"type":53,"value":229},"d",{"type":47,"tag":191,"props":231,"children":232},{"style":215},[233],{"type":53,"value":234},">",{"type":47,"tag":191,"props":236,"children":237},{"style":204},[238],{"type":53,"value":239}," --repo",{"type":47,"tag":191,"props":241,"children":242},{"style":204},[243],{"type":53,"value":244}," NVIDIA\u002FMegatron-LM",{"type":47,"tag":191,"props":246,"children":247},{"style":204},[248],{"type":53,"value":249}," --json",{"type":47,"tag":191,"props":251,"children":252},{"style":204},[253],{"type":53,"value":254}," jobs",{"type":47,"tag":191,"props":256,"children":257},{"style":226},[258],{"type":53,"value":259}," \\\n",{"type":47,"tag":191,"props":261,"children":263},{"class":193,"line":262},2,[264,269,274,279],{"type":47,"tag":191,"props":265,"children":266},{"style":204},[267],{"type":53,"value":268},"  --jq",{"type":47,"tag":191,"props":270,"children":271},{"style":215},[272],{"type":53,"value":273}," '",{"type":47,"tag":191,"props":275,"children":276},{"style":204},[277],{"type":53,"value":278},"[.jobs[] | select(.conclusion == \"failure\") | {id: .databaseId, name: .name, url: .url}]",{"type":47,"tag":191,"props":280,"children":281},{"style":215},[282],{"type":53,"value":283},"'\n",{"type":47,"tag":285,"props":286,"children":287},"br",{},[],{"type":53,"value":289},"If multiple jobs failed, ask the user which one to triage, or triage all of them if they say so.",{"type":47,"tag":78,"props":291,"children":293},{"id":292},"_3-fetch-the-failure-logs",[294],{"type":53,"value":295},"3. Fetch the failure logs",{"type":47,"tag":56,"props":297,"children":298},{},[299],{"type":53,"value":300},"For each failed job, retrieve the logs and narrow them down to the failure:",{"type":47,"tag":180,"props":302,"children":304},{"className":182,"code":303,"language":184,"meta":185,"style":185},"# Pull the raw log and keep only error-bearing lines\ngh api repos\u002FNVIDIA\u002FMegatron-LM\u002Factions\u002Fjobs\u002F\u003Cjob_id>\u002Flogs 2>&1 \\\n  | grep -E \"(FAILED|ERROR|\\bError\\b|assert|Traceback|Exception|##\\[error\\])\" \\\n  | head -200\n",[305],{"type":47,"tag":62,"props":306,"children":307},{"__ignoreMap":185},[308,317,366,404],{"type":47,"tag":191,"props":309,"children":310},{"class":193,"line":194},[311],{"type":47,"tag":191,"props":312,"children":314},{"style":313},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[315],{"type":53,"value":316},"# Pull the raw log and keep only error-bearing lines\n",{"type":47,"tag":191,"props":318,"children":319},{"class":193,"line":262},[320,324,329,334,339,344,348,352,357,362],{"type":47,"tag":191,"props":321,"children":322},{"style":198},[323],{"type":53,"value":201},{"type":47,"tag":191,"props":325,"children":326},{"style":204},[327],{"type":53,"value":328}," api",{"type":47,"tag":191,"props":330,"children":331},{"style":204},[332],{"type":53,"value":333}," repos\u002FNVIDIA\u002FMegatron-LM\u002Factions\u002Fjobs\u002F",{"type":47,"tag":191,"props":335,"children":336},{"style":215},[337],{"type":53,"value":338},"\u003C",{"type":47,"tag":191,"props":340,"children":341},{"style":204},[342],{"type":53,"value":343},"job_i",{"type":47,"tag":191,"props":345,"children":346},{"style":226},[347],{"type":53,"value":229},{"type":47,"tag":191,"props":349,"children":350},{"style":215},[351],{"type":53,"value":234},{"type":47,"tag":191,"props":353,"children":354},{"style":204},[355],{"type":53,"value":356},"\u002Flogs",{"type":47,"tag":191,"props":358,"children":359},{"style":215},[360],{"type":53,"value":361}," 2>&1",{"type":47,"tag":191,"props":363,"children":364},{"style":226},[365],{"type":53,"value":259},{"type":47,"tag":191,"props":367,"children":369},{"class":193,"line":368},3,[370,375,380,385,390,395,400],{"type":47,"tag":191,"props":371,"children":372},{"style":215},[373],{"type":53,"value":374},"  |",{"type":47,"tag":191,"props":376,"children":377},{"style":198},[378],{"type":53,"value":379}," grep",{"type":47,"tag":191,"props":381,"children":382},{"style":204},[383],{"type":53,"value":384}," -E",{"type":47,"tag":191,"props":386,"children":387},{"style":215},[388],{"type":53,"value":389}," \"",{"type":47,"tag":191,"props":391,"children":392},{"style":204},[393],{"type":53,"value":394},"(FAILED|ERROR|\\bError\\b|assert|Traceback|Exception|##\\[error\\])",{"type":47,"tag":191,"props":396,"children":397},{"style":215},[398],{"type":53,"value":399},"\"",{"type":47,"tag":191,"props":401,"children":402},{"style":226},[403],{"type":53,"value":259},{"type":47,"tag":191,"props":405,"children":407},{"class":193,"line":406},4,[408,412,417],{"type":47,"tag":191,"props":409,"children":410},{"style":215},[411],{"type":53,"value":374},{"type":47,"tag":191,"props":413,"children":414},{"style":198},[415],{"type":53,"value":416}," head",{"type":47,"tag":191,"props":418,"children":419},{"style":204},[420],{"type":53,"value":421}," -200\n",{"type":47,"tag":56,"props":423,"children":424},{},[425],{"type":53,"value":426},"Also capture the full job name:",{"type":47,"tag":180,"props":428,"children":430},{"className":182,"code":429,"language":184,"meta":185,"style":185},"gh run view --job \u003Cjob_id> --repo NVIDIA\u002FMegatron-LM --json name --jq .name\n",[431],{"type":47,"tag":62,"props":432,"children":433},{"__ignoreMap":185},[434],{"type":47,"tag":191,"props":435,"children":436},{"class":193,"line":194},[437,441,445,449,454,458,462,466,470,474,478,482,487,492],{"type":47,"tag":191,"props":438,"children":439},{"style":198},[440],{"type":53,"value":201},{"type":47,"tag":191,"props":442,"children":443},{"style":204},[444],{"type":53,"value":207},{"type":47,"tag":191,"props":446,"children":447},{"style":204},[448],{"type":53,"value":212},{"type":47,"tag":191,"props":450,"children":451},{"style":204},[452],{"type":53,"value":453}," --job",{"type":47,"tag":191,"props":455,"children":456},{"style":215},[457],{"type":53,"value":218},{"type":47,"tag":191,"props":459,"children":460},{"style":204},[461],{"type":53,"value":343},{"type":47,"tag":191,"props":463,"children":464},{"style":226},[465],{"type":53,"value":229},{"type":47,"tag":191,"props":467,"children":468},{"style":215},[469],{"type":53,"value":234},{"type":47,"tag":191,"props":471,"children":472},{"style":204},[473],{"type":53,"value":239},{"type":47,"tag":191,"props":475,"children":476},{"style":204},[477],{"type":53,"value":244},{"type":47,"tag":191,"props":479,"children":480},{"style":204},[481],{"type":53,"value":249},{"type":47,"tag":191,"props":483,"children":484},{"style":204},[485],{"type":53,"value":486}," name",{"type":47,"tag":191,"props":488,"children":489},{"style":204},[490],{"type":53,"value":491}," --jq",{"type":47,"tag":191,"props":493,"children":494},{"style":204},[495],{"type":53,"value":496}," .name\n",{"type":47,"tag":56,"props":498,"children":499},{},[500,502,508],{"type":53,"value":501},"If the grep output is sparse, download the full logs and look for the pytest\n",{"type":47,"tag":62,"props":503,"children":505},{"className":504},[],[506],{"type":53,"value":507},"FAILURES",{"type":53,"value":509}," section or the last non-zero exit signal.",{"type":47,"tag":78,"props":511,"children":513},{"id":512},"_4-resolve-the-triggering-pr-and-test-author",[514],{"type":53,"value":515},"4. Resolve the triggering PR and test author",{"type":47,"tag":56,"props":517,"children":518},{},[519,524,526,532],{"type":47,"tag":98,"props":520,"children":521},{},[522],{"type":53,"value":523},"Triggering PR",{"type":53,"value":525},": the run's head branch follows the pattern ",{"type":47,"tag":62,"props":527,"children":529},{"className":528},[],[530],{"type":53,"value":531},"pull-request\u002F\u003Cnumber>",{"type":53,"value":533},".\nExtract it and resolve the PR:",{"type":47,"tag":180,"props":535,"children":537},{"className":182,"code":536,"language":184,"meta":185,"style":185},"gh run view \u003Crun_id> --repo NVIDIA\u002FMegatron-LM --json headBranch --jq .headBranch\n# → e.g. \"pull-request\u002F4332\"\n# Extract PR number and fetch metadata:\ngh pr view \u003Cpr_number> --repo NVIDIA\u002FMegatron-LM --json number,title,url \\\n  --jq '{number: .number, title: .title, url: .url}'\n",[538],{"type":47,"tag":62,"props":539,"children":540},{"__ignoreMap":185},[541,598,606,614,669],{"type":47,"tag":191,"props":542,"children":543},{"class":193,"line":194},[544,548,552,556,560,564,568,572,576,580,584,589,593],{"type":47,"tag":191,"props":545,"children":546},{"style":198},[547],{"type":53,"value":201},{"type":47,"tag":191,"props":549,"children":550},{"style":204},[551],{"type":53,"value":207},{"type":47,"tag":191,"props":553,"children":554},{"style":204},[555],{"type":53,"value":212},{"type":47,"tag":191,"props":557,"children":558},{"style":215},[559],{"type":53,"value":218},{"type":47,"tag":191,"props":561,"children":562},{"style":204},[563],{"type":53,"value":223},{"type":47,"tag":191,"props":565,"children":566},{"style":226},[567],{"type":53,"value":229},{"type":47,"tag":191,"props":569,"children":570},{"style":215},[571],{"type":53,"value":234},{"type":47,"tag":191,"props":573,"children":574},{"style":204},[575],{"type":53,"value":239},{"type":47,"tag":191,"props":577,"children":578},{"style":204},[579],{"type":53,"value":244},{"type":47,"tag":191,"props":581,"children":582},{"style":204},[583],{"type":53,"value":249},{"type":47,"tag":191,"props":585,"children":586},{"style":204},[587],{"type":53,"value":588}," headBranch",{"type":47,"tag":191,"props":590,"children":591},{"style":204},[592],{"type":53,"value":491},{"type":47,"tag":191,"props":594,"children":595},{"style":204},[596],{"type":53,"value":597}," .headBranch\n",{"type":47,"tag":191,"props":599,"children":600},{"class":193,"line":262},[601],{"type":47,"tag":191,"props":602,"children":603},{"style":313},[604],{"type":53,"value":605},"# → e.g. \"pull-request\u002F4332\"\n",{"type":47,"tag":191,"props":607,"children":608},{"class":193,"line":368},[609],{"type":47,"tag":191,"props":610,"children":611},{"style":313},[612],{"type":53,"value":613},"# Extract PR number and fetch metadata:\n",{"type":47,"tag":191,"props":615,"children":616},{"class":193,"line":406},[617,621,626,630,634,639,644,648,652,656,660,665],{"type":47,"tag":191,"props":618,"children":619},{"style":198},[620],{"type":53,"value":201},{"type":47,"tag":191,"props":622,"children":623},{"style":204},[624],{"type":53,"value":625}," pr",{"type":47,"tag":191,"props":627,"children":628},{"style":204},[629],{"type":53,"value":212},{"type":47,"tag":191,"props":631,"children":632},{"style":215},[633],{"type":53,"value":218},{"type":47,"tag":191,"props":635,"children":636},{"style":204},[637],{"type":53,"value":638},"pr_numbe",{"type":47,"tag":191,"props":640,"children":641},{"style":226},[642],{"type":53,"value":643},"r",{"type":47,"tag":191,"props":645,"children":646},{"style":215},[647],{"type":53,"value":234},{"type":47,"tag":191,"props":649,"children":650},{"style":204},[651],{"type":53,"value":239},{"type":47,"tag":191,"props":653,"children":654},{"style":204},[655],{"type":53,"value":244},{"type":47,"tag":191,"props":657,"children":658},{"style":204},[659],{"type":53,"value":249},{"type":47,"tag":191,"props":661,"children":662},{"style":204},[663],{"type":53,"value":664}," number,title,url",{"type":47,"tag":191,"props":666,"children":667},{"style":226},[668],{"type":53,"value":259},{"type":47,"tag":191,"props":670,"children":672},{"class":193,"line":671},5,[673,677,681,686],{"type":47,"tag":191,"props":674,"children":675},{"style":204},[676],{"type":53,"value":268},{"type":47,"tag":191,"props":678,"children":679},{"style":215},[680],{"type":53,"value":273},{"type":47,"tag":191,"props":682,"children":683},{"style":204},[684],{"type":53,"value":685},"{number: .number, title: .title, url: .url}",{"type":47,"tag":191,"props":687,"children":688},{"style":215},[689],{"type":53,"value":283},{"type":47,"tag":56,"props":691,"children":692},{},[693,698,700,706],{"type":47,"tag":98,"props":694,"children":695},{},[696],{"type":53,"value":697},"Test file author",{"type":53,"value":699},": find the GitHub login of whoever last touched the failing\ntest file. The file may not exist on ",{"type":47,"tag":62,"props":701,"children":703},{"className":702},[],[704],{"type":53,"value":705},"main",{"type":53,"value":707}," — first determine the PR's base\nbranch, then search from there:",{"type":47,"tag":180,"props":709,"children":711},{"className":182,"code":710,"language":184,"meta":185,"style":185},"# 1. Get the PR's base branch (e.g. \"main\", \"dev\", \"release\u002FX.Y\")\ngh pr view \u003Cpr_number> --repo NVIDIA\u002FMegatron-LM --json baseRefName --jq .baseRefName\n\n# 2. Search commits on that base branch\ngh api \"repos\u002FNVIDIA\u002FMegatron-LM\u002Fcommits?path=\u003Ctest-file-path>&sha=\u003Cbase-branch>&per_page=1\" \\\n  --jq '.[0] | {login: .author.login, name: .commit.author.name, sha: .sha}'\n",[712],{"type":47,"tag":62,"props":713,"children":714},{"__ignoreMap":185},[715,723,780,788,796,824],{"type":47,"tag":191,"props":716,"children":717},{"class":193,"line":194},[718],{"type":47,"tag":191,"props":719,"children":720},{"style":313},[721],{"type":53,"value":722},"# 1. Get the PR's base branch (e.g. \"main\", \"dev\", \"release\u002FX.Y\")\n",{"type":47,"tag":191,"props":724,"children":725},{"class":193,"line":262},[726,730,734,738,742,746,750,754,758,762,766,771,775],{"type":47,"tag":191,"props":727,"children":728},{"style":198},[729],{"type":53,"value":201},{"type":47,"tag":191,"props":731,"children":732},{"style":204},[733],{"type":53,"value":625},{"type":47,"tag":191,"props":735,"children":736},{"style":204},[737],{"type":53,"value":212},{"type":47,"tag":191,"props":739,"children":740},{"style":215},[741],{"type":53,"value":218},{"type":47,"tag":191,"props":743,"children":744},{"style":204},[745],{"type":53,"value":638},{"type":47,"tag":191,"props":747,"children":748},{"style":226},[749],{"type":53,"value":643},{"type":47,"tag":191,"props":751,"children":752},{"style":215},[753],{"type":53,"value":234},{"type":47,"tag":191,"props":755,"children":756},{"style":204},[757],{"type":53,"value":239},{"type":47,"tag":191,"props":759,"children":760},{"style":204},[761],{"type":53,"value":244},{"type":47,"tag":191,"props":763,"children":764},{"style":204},[765],{"type":53,"value":249},{"type":47,"tag":191,"props":767,"children":768},{"style":204},[769],{"type":53,"value":770}," baseRefName",{"type":47,"tag":191,"props":772,"children":773},{"style":204},[774],{"type":53,"value":491},{"type":47,"tag":191,"props":776,"children":777},{"style":204},[778],{"type":53,"value":779}," .baseRefName\n",{"type":47,"tag":191,"props":781,"children":782},{"class":193,"line":368},[783],{"type":47,"tag":191,"props":784,"children":785},{"emptyLinePlaceholder":39},[786],{"type":53,"value":787},"\n",{"type":47,"tag":191,"props":789,"children":790},{"class":193,"line":406},[791],{"type":47,"tag":191,"props":792,"children":793},{"style":313},[794],{"type":53,"value":795},"# 2. Search commits on that base branch\n",{"type":47,"tag":191,"props":797,"children":798},{"class":193,"line":671},[799,803,807,811,816,820],{"type":47,"tag":191,"props":800,"children":801},{"style":198},[802],{"type":53,"value":201},{"type":47,"tag":191,"props":804,"children":805},{"style":204},[806],{"type":53,"value":328},{"type":47,"tag":191,"props":808,"children":809},{"style":215},[810],{"type":53,"value":389},{"type":47,"tag":191,"props":812,"children":813},{"style":204},[814],{"type":53,"value":815},"repos\u002FNVIDIA\u002FMegatron-LM\u002Fcommits?path=\u003Ctest-file-path>&sha=\u003Cbase-branch>&per_page=1",{"type":47,"tag":191,"props":817,"children":818},{"style":215},[819],{"type":53,"value":399},{"type":47,"tag":191,"props":821,"children":822},{"style":226},[823],{"type":53,"value":259},{"type":47,"tag":191,"props":825,"children":827},{"class":193,"line":826},6,[828,832,836,841],{"type":47,"tag":191,"props":829,"children":830},{"style":204},[831],{"type":53,"value":268},{"type":47,"tag":191,"props":833,"children":834},{"style":215},[835],{"type":53,"value":273},{"type":47,"tag":191,"props":837,"children":838},{"style":204},[839],{"type":53,"value":840},".[0] | {login: .author.login, name: .commit.author.name, sha: .sha}",{"type":47,"tag":191,"props":842,"children":843},{"style":215},[844],{"type":53,"value":283},{"type":47,"tag":56,"props":846,"children":847},{},[848],{"type":53,"value":849},"If the result is empty (file was introduced by the PR itself), query the PR's\ncommits instead:",{"type":47,"tag":180,"props":851,"children":853},{"className":182,"code":852,"language":184,"meta":185,"style":185},"gh api \"repos\u002FNVIDIA\u002FMegatron-LM\u002Fpulls\u002F\u003Cpr_number>\u002Fcommits\" \\\n  --jq '[.[] | select(.files? \u002F\u002F [] | any(.filename == \"\u003Ctest-file-path>\"))] | .[0].author.login'\n",[854],{"type":47,"tag":62,"props":855,"children":856},{"__ignoreMap":185},[857,885],{"type":47,"tag":191,"props":858,"children":859},{"class":193,"line":194},[860,864,868,872,877,881],{"type":47,"tag":191,"props":861,"children":862},{"style":198},[863],{"type":53,"value":201},{"type":47,"tag":191,"props":865,"children":866},{"style":204},[867],{"type":53,"value":328},{"type":47,"tag":191,"props":869,"children":870},{"style":215},[871],{"type":53,"value":389},{"type":47,"tag":191,"props":873,"children":874},{"style":204},[875],{"type":53,"value":876},"repos\u002FNVIDIA\u002FMegatron-LM\u002Fpulls\u002F\u003Cpr_number>\u002Fcommits",{"type":47,"tag":191,"props":878,"children":879},{"style":215},[880],{"type":53,"value":399},{"type":47,"tag":191,"props":882,"children":883},{"style":226},[884],{"type":53,"value":259},{"type":47,"tag":191,"props":886,"children":887},{"class":193,"line":262},[888,892,896,901],{"type":47,"tag":191,"props":889,"children":890},{"style":204},[891],{"type":53,"value":268},{"type":47,"tag":191,"props":893,"children":894},{"style":215},[895],{"type":53,"value":273},{"type":47,"tag":191,"props":897,"children":898},{"style":204},[899],{"type":53,"value":900},"[.[] | select(.files? \u002F\u002F [] | any(.filename == \"\u003Ctest-file-path>\"))] | .[0].author.login",{"type":47,"tag":191,"props":902,"children":903},{"style":215},[904],{"type":53,"value":283},{"type":47,"tag":56,"props":906,"children":907},{},[908],{"type":53,"value":909},"As a last resort, list the PR commits and pick the author of the commit whose\nmessage most closely relates to the failing test file.",{"type":47,"tag":78,"props":911,"children":913},{"id":912},"_5-extract-the-root-cause",[914],{"type":53,"value":915},"5. Extract the root cause",{"type":47,"tag":56,"props":917,"children":918},{},[919],{"type":53,"value":920},"From the logs, identify:",{"type":47,"tag":90,"props":922,"children":923},{},[924,942,952,970],{"type":47,"tag":94,"props":925,"children":926},{},[927,932,934,940],{"type":47,"tag":98,"props":928,"children":929},{},[930],{"type":53,"value":931},"Failed test(s)",{"type":53,"value":933},": lines matching ",{"type":47,"tag":62,"props":935,"children":937},{"className":936},[],[938],{"type":53,"value":939},"FAILED tests\u002F...::...",{"type":53,"value":941}," give the exact pytest node IDs.",{"type":47,"tag":94,"props":943,"children":944},{},[945,950],{"type":47,"tag":98,"props":946,"children":947},{},[948],{"type":53,"value":949},"Error message",{"type":53,"value":951},": the assertion failure, exception type, or first meaningful\ntraceback frame — keep it under ~30 lines.",{"type":47,"tag":94,"props":953,"children":954},{},[955,960,962,968],{"type":47,"tag":98,"props":956,"children":957},{},[958],{"type":53,"value":959},"Job name",{"type":53,"value":961},": the GitHub Actions job name (e.g. ",{"type":47,"tag":62,"props":963,"children":965},{"className":964},[],[966],{"type":53,"value":967},"tests\u002Funit_tests\u002Ftransformer\u002Fmoe\u002F**\u002F*.py - latest",{"type":53,"value":969},").",{"type":47,"tag":94,"props":971,"children":972},{},[973,978,980,985],{"type":47,"tag":98,"props":974,"children":975},{},[976],{"type":53,"value":977},"Run \u002F job URLs",{"type":53,"value":979}," and ",{"type":47,"tag":98,"props":981,"children":982},{},[983],{"type":53,"value":984},"PR URL",{"type":53,"value":986},": for linking in the issue.",{"type":47,"tag":78,"props":988,"children":990},{"id":989},"_6-check-for-duplicate-issues",[991],{"type":53,"value":992},"6. Check for duplicate issues",{"type":47,"tag":56,"props":994,"children":995},{},[996],{"type":53,"value":997},"Search for open issues that already cover the same test:",{"type":47,"tag":180,"props":999,"children":1001},{"className":182,"code":1000,"language":184,"meta":185,"style":185},"gh issue list --repo NVIDIA\u002FMegatron-LM \\\n  --state open \\\n  --search \"\u003Cfailed-test-filename>\" \\\n  --json number,title,url \\\n  --limit 10\n",[1002],{"type":47,"tag":62,"props":1003,"children":1004},{"__ignoreMap":185},[1005,1034,1051,1076,1092],{"type":47,"tag":191,"props":1006,"children":1007},{"class":193,"line":194},[1008,1012,1017,1022,1026,1030],{"type":47,"tag":191,"props":1009,"children":1010},{"style":198},[1011],{"type":53,"value":201},{"type":47,"tag":191,"props":1013,"children":1014},{"style":204},[1015],{"type":53,"value":1016}," issue",{"type":47,"tag":191,"props":1018,"children":1019},{"style":204},[1020],{"type":53,"value":1021}," list",{"type":47,"tag":191,"props":1023,"children":1024},{"style":204},[1025],{"type":53,"value":239},{"type":47,"tag":191,"props":1027,"children":1028},{"style":204},[1029],{"type":53,"value":244},{"type":47,"tag":191,"props":1031,"children":1032},{"style":226},[1033],{"type":53,"value":259},{"type":47,"tag":191,"props":1035,"children":1036},{"class":193,"line":262},[1037,1042,1047],{"type":47,"tag":191,"props":1038,"children":1039},{"style":204},[1040],{"type":53,"value":1041},"  --state",{"type":47,"tag":191,"props":1043,"children":1044},{"style":204},[1045],{"type":53,"value":1046}," open",{"type":47,"tag":191,"props":1048,"children":1049},{"style":226},[1050],{"type":53,"value":259},{"type":47,"tag":191,"props":1052,"children":1053},{"class":193,"line":368},[1054,1059,1063,1068,1072],{"type":47,"tag":191,"props":1055,"children":1056},{"style":204},[1057],{"type":53,"value":1058},"  --search",{"type":47,"tag":191,"props":1060,"children":1061},{"style":215},[1062],{"type":53,"value":389},{"type":47,"tag":191,"props":1064,"children":1065},{"style":204},[1066],{"type":53,"value":1067},"\u003Cfailed-test-filename>",{"type":47,"tag":191,"props":1069,"children":1070},{"style":215},[1071],{"type":53,"value":399},{"type":47,"tag":191,"props":1073,"children":1074},{"style":226},[1075],{"type":53,"value":259},{"type":47,"tag":191,"props":1077,"children":1078},{"class":193,"line":406},[1079,1084,1088],{"type":47,"tag":191,"props":1080,"children":1081},{"style":204},[1082],{"type":53,"value":1083},"  --json",{"type":47,"tag":191,"props":1085,"children":1086},{"style":204},[1087],{"type":53,"value":664},{"type":47,"tag":191,"props":1089,"children":1090},{"style":226},[1091],{"type":53,"value":259},{"type":47,"tag":191,"props":1093,"children":1094},{"class":193,"line":671},[1095,1100],{"type":47,"tag":191,"props":1096,"children":1097},{"style":204},[1098],{"type":53,"value":1099},"  --limit",{"type":47,"tag":191,"props":1101,"children":1103},{"style":1102},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[1104],{"type":53,"value":1105}," 10\n",{"type":47,"tag":90,"props":1107,"children":1108},{},[1109,1121],{"type":47,"tag":94,"props":1110,"children":1111},{},[1112,1114,1119],{"type":53,"value":1113},"If a matching open issue exists, ",{"type":47,"tag":98,"props":1115,"children":1116},{},[1117],{"type":53,"value":1118},"do not create a new one",{"type":53,"value":1120},". Report the\nexisting issue to the user and stop.",{"type":47,"tag":94,"props":1122,"children":1123},{},[1124],{"type":53,"value":1125},"If no match is found, proceed to file a new issue.",{"type":47,"tag":78,"props":1127,"children":1129},{"id":1128},"_7-create-the-issue",[1130],{"type":53,"value":1131},"7. Create the issue",{"type":47,"tag":56,"props":1133,"children":1134},{},[1135,1137,1143],{"type":53,"value":1136},"Pass ",{"type":47,"tag":62,"props":1138,"children":1140},{"className":1139},[],[1141],{"type":53,"value":1142},"--assignee \u003Ctest-author-login>",{"type":53,"value":1144}," to assign the issue to the test file's\nauthor. Include the triggering PR URL in the issue body.",{"type":47,"tag":180,"props":1146,"children":1148},{"className":182,"code":1147,"language":184,"meta":185,"style":185},"gh issue create \\\n  --repo NVIDIA\u002FMegatron-LM \\\n  --title \"🐛 CI failure: \u003Cfailed-test-node-id>\" \\\n  --label \"bug\" \\\n  --assignee \"\u003Ctest-author-login>\" \\\n  --body \"...\"\n",[1149],{"type":47,"tag":62,"props":1150,"children":1151},{"__ignoreMap":185},[1152,1172,1188,1213,1238,1263],{"type":47,"tag":191,"props":1153,"children":1154},{"class":193,"line":194},[1155,1159,1163,1168],{"type":47,"tag":191,"props":1156,"children":1157},{"style":198},[1158],{"type":53,"value":201},{"type":47,"tag":191,"props":1160,"children":1161},{"style":204},[1162],{"type":53,"value":1016},{"type":47,"tag":191,"props":1164,"children":1165},{"style":204},[1166],{"type":53,"value":1167}," create",{"type":47,"tag":191,"props":1169,"children":1170},{"style":226},[1171],{"type":53,"value":259},{"type":47,"tag":191,"props":1173,"children":1174},{"class":193,"line":262},[1175,1180,1184],{"type":47,"tag":191,"props":1176,"children":1177},{"style":204},[1178],{"type":53,"value":1179},"  --repo",{"type":47,"tag":191,"props":1181,"children":1182},{"style":204},[1183],{"type":53,"value":244},{"type":47,"tag":191,"props":1185,"children":1186},{"style":226},[1187],{"type":53,"value":259},{"type":47,"tag":191,"props":1189,"children":1190},{"class":193,"line":368},[1191,1196,1200,1205,1209],{"type":47,"tag":191,"props":1192,"children":1193},{"style":204},[1194],{"type":53,"value":1195},"  --title",{"type":47,"tag":191,"props":1197,"children":1198},{"style":215},[1199],{"type":53,"value":389},{"type":47,"tag":191,"props":1201,"children":1202},{"style":204},[1203],{"type":53,"value":1204},"🐛 CI failure: \u003Cfailed-test-node-id>",{"type":47,"tag":191,"props":1206,"children":1207},{"style":215},[1208],{"type":53,"value":399},{"type":47,"tag":191,"props":1210,"children":1211},{"style":226},[1212],{"type":53,"value":259},{"type":47,"tag":191,"props":1214,"children":1215},{"class":193,"line":406},[1216,1221,1225,1230,1234],{"type":47,"tag":191,"props":1217,"children":1218},{"style":204},[1219],{"type":53,"value":1220},"  --label",{"type":47,"tag":191,"props":1222,"children":1223},{"style":215},[1224],{"type":53,"value":389},{"type":47,"tag":191,"props":1226,"children":1227},{"style":204},[1228],{"type":53,"value":1229},"bug",{"type":47,"tag":191,"props":1231,"children":1232},{"style":215},[1233],{"type":53,"value":399},{"type":47,"tag":191,"props":1235,"children":1236},{"style":226},[1237],{"type":53,"value":259},{"type":47,"tag":191,"props":1239,"children":1240},{"class":193,"line":671},[1241,1246,1250,1255,1259],{"type":47,"tag":191,"props":1242,"children":1243},{"style":204},[1244],{"type":53,"value":1245},"  --assignee",{"type":47,"tag":191,"props":1247,"children":1248},{"style":215},[1249],{"type":53,"value":389},{"type":47,"tag":191,"props":1251,"children":1252},{"style":204},[1253],{"type":53,"value":1254},"\u003Ctest-author-login>",{"type":47,"tag":191,"props":1256,"children":1257},{"style":215},[1258],{"type":53,"value":399},{"type":47,"tag":191,"props":1260,"children":1261},{"style":226},[1262],{"type":53,"value":259},{"type":47,"tag":191,"props":1264,"children":1265},{"class":193,"line":826},[1266,1271,1275,1280],{"type":47,"tag":191,"props":1267,"children":1268},{"style":204},[1269],{"type":53,"value":1270},"  --body",{"type":47,"tag":191,"props":1272,"children":1273},{"style":215},[1274],{"type":53,"value":389},{"type":47,"tag":191,"props":1276,"children":1277},{"style":204},[1278],{"type":53,"value":1279},"...",{"type":47,"tag":191,"props":1281,"children":1282},{"style":215},[1283],{"type":53,"value":1284},"\"\n",{"type":47,"tag":56,"props":1286,"children":1287},{},[1288],{"type":53,"value":1289},"Use the bug-report template body structure:",{"type":47,"tag":180,"props":1291,"children":1295},{"className":1292,"code":1293,"language":1294,"meta":185,"style":185},"language-markdown shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","**Describe the bug**\n\nCI test `\u003Cfailed-test-node-id>` failed in job [`\u003Cjob-name>`](\u003Cjob-url>).\nTag @NVIDIA\u002Fmcore-oncall to get oncall's attention to this issue.\n\n**Failing run**\n\n| Field | Value |\n|-------|-------|\n| PR    | [#\u003Cpr_number>: \u003Cpr_title>](\u003Cpr_url>) |\n| Run   | [\u003Crun_id>](\u003Crun_url>) |\n| Job   | [\u003Cjob_name>](\u003Cjob_url>) |\n\n**Error**\n\n","markdown",[1296],{"type":47,"tag":62,"props":1297,"children":1298},{"__ignoreMap":185},[1299,1319,1326,1392,1400,1407,1423,1431,1459,1468,1513,1556,1599,1607],{"type":47,"tag":191,"props":1300,"children":1301},{"class":193,"line":194},[1302,1308,1314],{"type":47,"tag":191,"props":1303,"children":1305},{"style":1304},"--shiki-light:#39ADB5;--shiki-light-font-weight:bold;--shiki-default:#89DDFF;--shiki-default-font-weight:bold;--shiki-dark:#89DDFF;--shiki-dark-font-weight:bold",[1306],{"type":53,"value":1307},"**",{"type":47,"tag":191,"props":1309,"children":1311},{"style":1310},"--shiki-light:#E53935;--shiki-light-font-weight:bold;--shiki-default:#F07178;--shiki-default-font-weight:bold;--shiki-dark:#F07178;--shiki-dark-font-weight:bold",[1312],{"type":53,"value":1313},"Describe the bug",{"type":47,"tag":191,"props":1315,"children":1316},{"style":1304},[1317],{"type":53,"value":1318},"**\n",{"type":47,"tag":191,"props":1320,"children":1321},{"class":193,"line":262},[1322],{"type":47,"tag":191,"props":1323,"children":1324},{"emptyLinePlaceholder":39},[1325],{"type":53,"value":787},{"type":47,"tag":191,"props":1327,"children":1328},{"class":193,"line":368},[1329,1334,1339,1344,1348,1353,1358,1362,1367,1371,1376,1382,1387],{"type":47,"tag":191,"props":1330,"children":1331},{"style":226},[1332],{"type":53,"value":1333},"CI test ",{"type":47,"tag":191,"props":1335,"children":1336},{"style":215},[1337],{"type":53,"value":1338},"`",{"type":47,"tag":191,"props":1340,"children":1341},{"style":204},[1342],{"type":53,"value":1343},"\u003Cfailed-test-node-id>",{"type":47,"tag":191,"props":1345,"children":1346},{"style":215},[1347],{"type":53,"value":1338},{"type":47,"tag":191,"props":1349,"children":1350},{"style":226},[1351],{"type":53,"value":1352}," failed in job ",{"type":47,"tag":191,"props":1354,"children":1355},{"style":215},[1356],{"type":53,"value":1357},"[",{"type":47,"tag":191,"props":1359,"children":1360},{"style":215},[1361],{"type":53,"value":1338},{"type":47,"tag":191,"props":1363,"children":1364},{"style":204},[1365],{"type":53,"value":1366},"\u003Cjob-name>",{"type":47,"tag":191,"props":1368,"children":1369},{"style":215},[1370],{"type":53,"value":1338},{"type":47,"tag":191,"props":1372,"children":1373},{"style":215},[1374],{"type":53,"value":1375},"](\u003C",{"type":47,"tag":191,"props":1377,"children":1379},{"style":1378},"--shiki-light:#E53935;--shiki-light-text-decoration:underline;--shiki-default:#F07178;--shiki-default-text-decoration:underline;--shiki-dark:#F07178;--shiki-dark-text-decoration:underline",[1380],{"type":53,"value":1381},"job-url",{"type":47,"tag":191,"props":1383,"children":1384},{"style":215},[1385],{"type":53,"value":1386},">)",{"type":47,"tag":191,"props":1388,"children":1389},{"style":226},[1390],{"type":53,"value":1391},".\n",{"type":47,"tag":191,"props":1393,"children":1394},{"class":193,"line":406},[1395],{"type":47,"tag":191,"props":1396,"children":1397},{"style":226},[1398],{"type":53,"value":1399},"Tag @NVIDIA\u002Fmcore-oncall to get oncall's attention to this issue.\n",{"type":47,"tag":191,"props":1401,"children":1402},{"class":193,"line":671},[1403],{"type":47,"tag":191,"props":1404,"children":1405},{"emptyLinePlaceholder":39},[1406],{"type":53,"value":787},{"type":47,"tag":191,"props":1408,"children":1409},{"class":193,"line":826},[1410,1414,1419],{"type":47,"tag":191,"props":1411,"children":1412},{"style":1304},[1413],{"type":53,"value":1307},{"type":47,"tag":191,"props":1415,"children":1416},{"style":1310},[1417],{"type":53,"value":1418},"Failing run",{"type":47,"tag":191,"props":1420,"children":1421},{"style":1304},[1422],{"type":53,"value":1318},{"type":47,"tag":191,"props":1424,"children":1426},{"class":193,"line":1425},7,[1427],{"type":47,"tag":191,"props":1428,"children":1429},{"emptyLinePlaceholder":39},[1430],{"type":53,"value":787},{"type":47,"tag":191,"props":1432,"children":1434},{"class":193,"line":1433},8,[1435,1440,1445,1449,1454],{"type":47,"tag":191,"props":1436,"children":1437},{"style":215},[1438],{"type":53,"value":1439},"|",{"type":47,"tag":191,"props":1441,"children":1442},{"style":226},[1443],{"type":53,"value":1444}," Field ",{"type":47,"tag":191,"props":1446,"children":1447},{"style":215},[1448],{"type":53,"value":1439},{"type":47,"tag":191,"props":1450,"children":1451},{"style":226},[1452],{"type":53,"value":1453}," Value ",{"type":47,"tag":191,"props":1455,"children":1456},{"style":215},[1457],{"type":53,"value":1458},"|\n",{"type":47,"tag":191,"props":1460,"children":1462},{"class":193,"line":1461},9,[1463],{"type":47,"tag":191,"props":1464,"children":1465},{"style":215},[1466],{"type":53,"value":1467},"|-------|-------|\n",{"type":47,"tag":191,"props":1469,"children":1471},{"class":193,"line":1470},10,[1472,1476,1481,1485,1490,1495,1499,1504,1508],{"type":47,"tag":191,"props":1473,"children":1474},{"style":215},[1475],{"type":53,"value":1439},{"type":47,"tag":191,"props":1477,"children":1478},{"style":226},[1479],{"type":53,"value":1480}," PR    ",{"type":47,"tag":191,"props":1482,"children":1483},{"style":215},[1484],{"type":53,"value":1439},{"type":47,"tag":191,"props":1486,"children":1487},{"style":215},[1488],{"type":53,"value":1489}," [",{"type":47,"tag":191,"props":1491,"children":1492},{"style":204},[1493],{"type":53,"value":1494},"#\u003Cpr_number>: \u003Cpr_title>",{"type":47,"tag":191,"props":1496,"children":1497},{"style":215},[1498],{"type":53,"value":1375},{"type":47,"tag":191,"props":1500,"children":1501},{"style":1378},[1502],{"type":53,"value":1503},"pr_url",{"type":47,"tag":191,"props":1505,"children":1506},{"style":215},[1507],{"type":53,"value":1386},{"type":47,"tag":191,"props":1509,"children":1510},{"style":215},[1511],{"type":53,"value":1512}," |\n",{"type":47,"tag":191,"props":1514,"children":1516},{"class":193,"line":1515},11,[1517,1521,1526,1530,1534,1539,1543,1548,1552],{"type":47,"tag":191,"props":1518,"children":1519},{"style":215},[1520],{"type":53,"value":1439},{"type":47,"tag":191,"props":1522,"children":1523},{"style":226},[1524],{"type":53,"value":1525}," Run   ",{"type":47,"tag":191,"props":1527,"children":1528},{"style":215},[1529],{"type":53,"value":1439},{"type":47,"tag":191,"props":1531,"children":1532},{"style":215},[1533],{"type":53,"value":1489},{"type":47,"tag":191,"props":1535,"children":1536},{"style":204},[1537],{"type":53,"value":1538},"\u003Crun_id>",{"type":47,"tag":191,"props":1540,"children":1541},{"style":215},[1542],{"type":53,"value":1375},{"type":47,"tag":191,"props":1544,"children":1545},{"style":1378},[1546],{"type":53,"value":1547},"run_url",{"type":47,"tag":191,"props":1549,"children":1550},{"style":215},[1551],{"type":53,"value":1386},{"type":47,"tag":191,"props":1553,"children":1554},{"style":215},[1555],{"type":53,"value":1512},{"type":47,"tag":191,"props":1557,"children":1559},{"class":193,"line":1558},12,[1560,1564,1569,1573,1577,1582,1586,1591,1595],{"type":47,"tag":191,"props":1561,"children":1562},{"style":215},[1563],{"type":53,"value":1439},{"type":47,"tag":191,"props":1565,"children":1566},{"style":226},[1567],{"type":53,"value":1568}," Job   ",{"type":47,"tag":191,"props":1570,"children":1571},{"style":215},[1572],{"type":53,"value":1439},{"type":47,"tag":191,"props":1574,"children":1575},{"style":215},[1576],{"type":53,"value":1489},{"type":47,"tag":191,"props":1578,"children":1579},{"style":204},[1580],{"type":53,"value":1581},"\u003Cjob_name>",{"type":47,"tag":191,"props":1583,"children":1584},{"style":215},[1585],{"type":53,"value":1375},{"type":47,"tag":191,"props":1587,"children":1588},{"style":1378},[1589],{"type":53,"value":1590},"job_url",{"type":47,"tag":191,"props":1592,"children":1593},{"style":215},[1594],{"type":53,"value":1386},{"type":47,"tag":191,"props":1596,"children":1597},{"style":215},[1598],{"type":53,"value":1512},{"type":47,"tag":191,"props":1600,"children":1602},{"class":193,"line":1601},13,[1603],{"type":47,"tag":191,"props":1604,"children":1605},{"emptyLinePlaceholder":39},[1606],{"type":53,"value":787},{"type":47,"tag":191,"props":1608,"children":1610},{"class":193,"line":1609},14,[1611,1615,1620],{"type":47,"tag":191,"props":1612,"children":1613},{"style":1304},[1614],{"type":53,"value":1307},{"type":47,"tag":191,"props":1616,"children":1617},{"style":1310},[1618],{"type":53,"value":1619},"Error",{"type":47,"tag":191,"props":1621,"children":1622},{"style":1304},[1623],{"type":53,"value":1318},{"type":47,"tag":56,"props":1625,"children":1626},{},[1627],{"type":53,"value":1628},"\u003Ccore error message \u002F traceback — 30 lines max>",{"type":47,"tag":180,"props":1630,"children":1634},{"className":1631,"code":1633,"language":53},[1632],"language-text","\n**Steps\u002FCode to reproduce bug**\n\nRe-run the failing CI job linked above, or locally inside the dev container:\n\n```bash\npytest \u003Cfailed-test-node-id>\n",[1635],{"type":47,"tag":62,"props":1636,"children":1637},{"__ignoreMap":185},[1638],{"type":53,"value":1633},{"type":47,"tag":56,"props":1640,"children":1641},{},[1642],{"type":47,"tag":98,"props":1643,"children":1644},{},[1645],{"type":53,"value":1646},"Additional context",{"type":47,"tag":56,"props":1648,"children":1649},{},[1650,1652,1658],{"type":53,"value":1651},"Triaged automatically via ",{"type":47,"tag":62,"props":1653,"children":1655},{"className":1654},[],[1656],{"type":53,"value":1657},"\u002Ftriage-issue",{"type":53,"value":69},{"type":47,"tag":180,"props":1660,"children":1663},{"className":1661,"code":1662,"language":53},[1632],"\nIf multiple tests failed in the same job, list each one as a separate bullet\nunder \"Describe the bug\" and include the combined error snippets. Assign the\nissue to the author of whichever test file appears first in the failure list.\n\n### 8. Report back to the user\n\nPrint the URL of the newly created issue (or the duplicate, if found) so the\nuser can review or share it.\n\n## Important guidelines\n\n- Never create an issue if a duplicate already exists — link the existing one instead.\n- Always include the triggering PR link in the issue body.\n- Always assign the issue to the test file's most recent author. If the author\n  lookup fails (e.g. the commit was made by a bot or the login is unavailable),\n  skip `--assignee` and note it in the \"Additional context\" section.\n- Keep the error snippet concise (≤30 lines). Truncate long tracebacks and note that the full log is available via the job URL.\n- Do not guess the root cause — quote the actual log output verbatim.\n- If the job is still in progress or the logs are unavailable, say so and ask the user to retry once the run completes.\n",[1664],{"type":47,"tag":62,"props":1665,"children":1666},{"__ignoreMap":185},[1667],{"type":53,"value":1662},{"type":47,"tag":1669,"props":1670,"children":1671},"style",{},[1672],{"type":53,"value":1673},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1675,"total":1601},[1676,1692,1703,1713,1719,1732,1746],{"slug":1677,"name":1677,"fn":1678,"description":1679,"org":1680,"tags":1681,"stars":22,"repoUrl":23,"updatedAt":1691},"mcore-build-and-dependency","manage Megatron-LM development environments","Container-based dev environment setup and dependency management for Megatron-LM. Covers acquiring and launching the CI container, uv package management, and updating uv.lock.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1682,1685,1688],{"name":1683,"slug":1684,"type":15},"Containers","containers",{"name":1686,"slug":1687,"type":15},"Deployment","deployment",{"name":1689,"slug":1690,"type":15},"Python","python","2026-07-27T06:06:11.249662",{"slug":1693,"name":1693,"fn":1694,"description":1695,"org":1696,"tags":1697,"stars":22,"repoUrl":23,"updatedAt":1702},"mcore-bump-base-image","update NVIDIA PyTorch base images","Bump the NVIDIA PyTorch base image (`nvcr.io\u002Fnvidia\u002Fpytorch:YY.MM-py3`) used by Megatron-LM CI. Covers the two pin sites (GitHub CI in `docker\u002F.ngc_version.dev` and GitLab CI in `.gitlab\u002Fstages\u002F01.build.yml`), the post-bump CI loop (re-run functional tests, refresh golden values, mark broken tests), and the gotchas that bit PRs",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1698,1701],{"name":1699,"slug":1700,"type":15},"CI\u002FCD","ci-cd",{"name":1686,"slug":1687,"type":15},"2026-07-14T05:25:59.97109",{"slug":1704,"name":1704,"fn":1705,"description":1706,"org":1707,"tags":1708,"stars":22,"repoUrl":23,"updatedAt":1712},"mcore-cicd","manage CI\u002FCD pipelines for Megatron-LM","CI\u002FCD reference for Megatron-LM. Covers CI pipeline structure, PR scope labels, triggering internal GitLab CI (which force-pushes the current branch to a pull-request\u002FBRANCH ref — always dry-run and verify the destination first; never run against shared or protected branches), and CI failure investigation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1709,1710,1711],{"name":1699,"slug":1700,"type":15},{"name":1686,"slug":1687,"type":15},{"name":13,"slug":14,"type":15},"2026-07-27T06:06:12.278222",{"slug":4,"name":4,"fn":5,"description":6,"org":1714,"tags":1715,"stars":22,"repoUrl":23,"updatedAt":24},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1716,1717,1718],{"name":20,"slug":21,"type":15},{"name":13,"slug":14,"type":15},{"name":17,"slug":18,"type":15},{"slug":1720,"name":1720,"fn":1721,"description":1722,"org":1723,"tags":1724,"stars":22,"repoUrl":23,"updatedAt":1731},"mcore-linting-and-formatting","lint and format Megatron-LM code","Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1725,1728],{"name":1726,"slug":1727,"type":15},"Best Practices","best-practices",{"name":1729,"slug":1730,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1733,"name":1733,"fn":1734,"description":1735,"org":1736,"tags":1737,"stars":22,"repoUrl":23,"updatedAt":1745},"mcore-migrate-gpt-to-hybrid","migrate Megatron-LM models to HybridModel","Migration guide for moving Megatron Core GPTModel checkpoints, model providers, training commands, and layer mappings to HybridModel.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1738,1741,1744],{"name":1739,"slug":1740,"type":15},"Machine Learning","machine-learning",{"name":1742,"slug":1743,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1747,"name":1747,"fn":1748,"description":1749,"org":1750,"tags":1751,"stars":22,"repoUrl":23,"updatedAt":1758},"mcore-onboard-gb200-1node-tests","onboard functional tests for GB200","Onboard 1-node GitHub MR functional tests for GB200 from existing mr-scoped 2-node tests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1752,1755],{"name":1753,"slug":1754,"type":15},"QA","qa",{"name":1756,"slug":1757,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"items":1760,"total":1863},[1761,1779,1785,1790,1796,1802,1807,1813,1818,1829,1843,1852],{"slug":1762,"name":1762,"fn":1763,"description":1764,"org":1765,"tags":1766,"stars":1776,"repoUrl":1777,"updatedAt":1778},"nemoclaw-user-guide","retrieve NemoClaw documentation and configuration","Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1767,1770,1773],{"name":1768,"slug":1769,"type":15},"Documentation","documentation",{"name":1771,"slug":1772,"type":15},"MCP","mcp",{"name":1774,"slug":1775,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1677,"name":1677,"fn":1678,"description":1679,"org":1780,"tags":1781,"stars":22,"repoUrl":23,"updatedAt":1691},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1782,1783,1784],{"name":1683,"slug":1684,"type":15},{"name":1686,"slug":1687,"type":15},{"name":1689,"slug":1690,"type":15},{"slug":1693,"name":1693,"fn":1694,"description":1695,"org":1786,"tags":1787,"stars":22,"repoUrl":23,"updatedAt":1702},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1788,1789],{"name":1699,"slug":1700,"type":15},{"name":1686,"slug":1687,"type":15},{"slug":1704,"name":1704,"fn":1705,"description":1706,"org":1791,"tags":1792,"stars":22,"repoUrl":23,"updatedAt":1712},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1793,1794,1795],{"name":1699,"slug":1700,"type":15},{"name":1686,"slug":1687,"type":15},{"name":13,"slug":14,"type":15},{"slug":4,"name":4,"fn":5,"description":6,"org":1797,"tags":1798,"stars":22,"repoUrl":23,"updatedAt":24},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1799,1800,1801],{"name":20,"slug":21,"type":15},{"name":13,"slug":14,"type":15},{"name":17,"slug":18,"type":15},{"slug":1720,"name":1720,"fn":1721,"description":1722,"org":1803,"tags":1804,"stars":22,"repoUrl":23,"updatedAt":1731},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1805,1806],{"name":1726,"slug":1727,"type":15},{"name":1729,"slug":1730,"type":15},{"slug":1733,"name":1733,"fn":1734,"description":1735,"org":1808,"tags":1809,"stars":22,"repoUrl":23,"updatedAt":1745},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1810,1811,1812],{"name":1739,"slug":1740,"type":15},{"name":1742,"slug":1743,"type":15},{"name":9,"slug":8,"type":15},{"slug":1747,"name":1747,"fn":1748,"description":1749,"org":1814,"tags":1815,"stars":22,"repoUrl":23,"updatedAt":1758},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1816,1817],{"name":1753,"slug":1754,"type":15},{"name":1756,"slug":1757,"type":15},{"slug":1819,"name":1819,"fn":1820,"description":1821,"org":1822,"tags":1823,"stars":22,"repoUrl":23,"updatedAt":1828},"mcore-run-on-slurm","launch distributed training jobs on SLURM","How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1824,1825],{"name":1686,"slug":1687,"type":15},{"name":1826,"slug":1827,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":1830,"name":1830,"fn":1831,"description":1832,"org":1833,"tags":1834,"stars":22,"repoUrl":23,"updatedAt":1842},"mcore-split-pr","split pull requests to reduce review load","Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1835,1838,1839],{"name":1836,"slug":1837,"type":15},"Code Review","code-review",{"name":13,"slug":14,"type":15},{"name":1840,"slug":1841,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1844,"name":1844,"fn":1845,"description":1846,"org":1847,"tags":1848,"stars":22,"repoUrl":23,"updatedAt":1851},"mcore-testing","run and manage Megatron-LM tests","Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1849,1850],{"name":1753,"slug":1754,"type":15},{"name":1756,"slug":1757,"type":15},"2026-07-14T05:25:54.928983",{"slug":1853,"name":1853,"fn":1854,"description":1855,"org":1856,"tags":1857,"stars":22,"repoUrl":23,"updatedAt":1862},"nightly-sync","manage nightly main-to-dev sync workflows","Domain knowledge for the nightly main-to-dev sync workflow. Covers merge strategy, CI architecture, failure investigation, and known issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1858,1861],{"name":1859,"slug":1860,"type":15},"Automation","automation",{"name":1699,"slug":1700,"type":15},"2026-07-30T05:29:03.275638",496]