027b078915
Scheduled GitHub Actions check (runs on GitHub infra, not the dead local cron): for each app repo, compares the latest production deployment SHA to main and flags failed deploy runs. Opens one de-duplicated 'deploy-health' issue when something is off — the alarm that was missing when deploys silently broke. Agent-Session: cto deploy-health-monitor
71 lines
3.7 KiB
YAML
71 lines
3.7 KiB
YAML
name: Deploy Health Monitor
|
|
|
|
# Daily drift + deploy-failure check across all peptiderepo app repos.
|
|
# Runs on GitHub's scheduler (independent of the local board cron that died).
|
|
# Opens ONE de-duplicated 'deploy-health' issue here when something is off.
|
|
|
|
on:
|
|
schedule:
|
|
- cron: '0 13 * * *'
|
|
workflow_dispatch:
|
|
|
|
permissions:
|
|
contents: read
|
|
issues: write
|
|
actions: read
|
|
|
|
jobs:
|
|
check:
|
|
name: Check prod-vs-main drift + last deploy status
|
|
runs-on: ubuntu-latest
|
|
steps:
|
|
- uses: actions/github-script@v7
|
|
with:
|
|
script: |
|
|
const owner = 'peptiderepo';
|
|
const repos = [
|
|
'prautoblogger','peptide-repo-core','peptide-search-ai',
|
|
'peptide-news-plugin','peptide-reconstitution-calculator','peptide-starter-theme'
|
|
];
|
|
const short = s => s ? String(s).substring(0,7) : '—';
|
|
const problems = [];
|
|
for (const repo of repos) {
|
|
try {
|
|
const main = (await github.rest.repos.getBranch({owner, repo, branch:'main'})).data.commit.sha;
|
|
let runConc = 'none', runSha = '';
|
|
try {
|
|
const runs = (await github.rest.actions.listWorkflowRuns({owner, repo, workflow_id:'deploy.yml', per_page:1})).data.workflow_runs;
|
|
if (runs.length) { runConc = runs[0].conclusion || runs[0].status; runSha = runs[0].head_sha; }
|
|
} catch (e) { runConc = 'no-deploy-workflow'; }
|
|
let prodSha = null, prodState = 'none';
|
|
const deps = (await github.rest.repos.listDeployments({owner, repo, environment:'production', per_page:1})).data;
|
|
if (deps.length) {
|
|
prodSha = deps[0].sha;
|
|
const sts = (await github.rest.repos.listDeploymentStatuses({owner, repo, deployment_id: deps[0].id, per_page:1})).data;
|
|
if (sts.length) prodState = sts[0].state;
|
|
}
|
|
if (runConc === 'failure') problems.push(`**${repo}**: latest deploy run failed (\`${short(runSha)}\`)`);
|
|
if (prodSha && prodSha !== main) problems.push(`**${repo}**: prod behind main — prod=\`${short(prodSha)}\` vs main=\`${short(main)}\` (undeployed commits)`);
|
|
if (prodSha && prodState !== 'success') problems.push(`**${repo}**: last production deployment state is \`${prodState}\``);
|
|
core.info(`${repo}: main=${short(main)} lastRun=${runConc} prodDeploy=${short(prodSha)}/${prodState}`);
|
|
} catch (e) {
|
|
problems.push(`**${repo}**: monitor error — ${e.message}`);
|
|
}
|
|
}
|
|
const summary = problems.length
|
|
? `### ⚠️ Deploy health: ${problems.length} issue(s)\n\n` + problems.map(p=>`- ${p}`).join('\n')
|
|
: `### ✅ All app repos current; last deploy healthy`;
|
|
await core.summary.addRaw(summary).write();
|
|
if (problems.length) {
|
|
const existing = (await github.rest.issues.listForRepo({owner, repo: context.repo.repo, labels:'deploy-health', state:'open'})).data;
|
|
if (existing.length === 0) {
|
|
await github.rest.issues.create({ owner, repo: context.repo.repo,
|
|
title: `Deploy health: ${problems.length} issue(s) detected`,
|
|
body: summary + `\n\n_Opened by deploy-health.yml. Close once resolved; the monitor won't reopen while this stays open._`,
|
|
labels: ['deploy-health'] });
|
|
core.warning(`Opened deploy-health issue (${problems.length} problems)`);
|
|
} else {
|
|
core.warning(`${problems.length} problems; deploy-health issue already open (#${existing[0].number})`);
|
|
}
|
|
}
|