badgr-cli 1.0.18 → 1.0.20

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "badgr-cli",
-  "version": "1.0.18",
+  "version": "1.0.20",
   "description": "Badgr, run or serve GPU workloads from one command",
   "type": "module",
   "bin": {

package/src/commands/run.js

@@ -18,6 +18,7 @@ export function parseRunArgs(args) {
     if (args[i] === '--image')          { flags.image = args[++i]; i++; continue; }
     if (args[i] === '--count')          { flags.count = parseInt(args[++i], 10); i++; continue; }
     if (args[i] === '--region')         { flags.region = args[++i]; i++; continue; }
+    if (args[i] === '--tier')           { flags.tier = args[++i]; i++; continue; }
     if (args[i] === '--max-price')      { flags.maxPrice = parseFloat(args[++i]); i++; continue; }
     if (args[i] === '--name')           { flags.name = args[++i]; i++; continue; }
     if (args[i] === '--detach')          { flags.detach = true; i++; continue; }
@@ -81,6 +82,42 @@ function renderStatusBar(chalk, { elapsedMs, ratePerHour, gpuUtil, cpuUtil, maxR
   return chalk.dim('  ' + parts.join('  •  '));
 }
 
+// Wait for status to leave 'starting' (smoke check running on backend).
+// Returns the final dep object with status 'running' or 'failed'.
+async function waitForRunning(config, depId, chalk) {
+  const POLL_MS = 3000;
+  const TIMEOUT_MS = 120_000; // 2 min max for smoke check
+  const startMs = Date.now();
+  let dots = 0;
+
+  process.stdout.write(chalk.dim('  Waiting for container to start'));
+  const ticker = setInterval(() => {
+    process.stdout.write('.');
+    dots++;
+  }, 1000);
+
+  try {
+    while (Date.now() - startMs < TIMEOUT_MS) {
+      await new Promise(r => setTimeout(r, POLL_MS));
+      const dep = await callApi(`/deployments/${depId}`, {
+        apiKey: config.apiKey,
+        baseUrl: config.baseUrl,
+      });
+      if (dep.status !== 'starting') {
+        clearInterval(ticker);
+        process.stdout.write('\n');
+        return dep;
+      }
+    }
+  } finally {
+    clearInterval(ticker);
+    if (dots > 0) process.stdout.write('\n');
+  }
+
+  // Timed out waiting — return whatever we have
+  return await callApi(`/deployments/${depId}`, { apiKey: config.apiKey, baseUrl: config.baseUrl });
+}
+
 async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost = null, ratePerHour = 0, onTeardown }) {
   const TERMINAL = new Set(['stopped', 'failed', 'completed']);
   const POLL_MS  = 4000;
@@ -129,14 +166,14 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
       if (maxCost !== null && spentSoFar >= maxCost) {
         tearing = true;
         stopTicker();
-        onTeardown('max-cost');
+        await onTeardown('max-cost');
         return { status: 'timeout', exitCode: null, runtimeMs: elapsedMs, failureType: null };
       }
 
       if (maxRuntimeMs !== null && elapsedMs >= maxRuntimeMs) {
         tearing = true;
         stopTicker();
-        onTeardown('max-runtime');
+        await onTeardown('max-runtime');
         return { status: 'timeout', exitCode: null, runtimeMs: elapsedMs, failureType: null };
       }
 
@@ -158,7 +195,7 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
           } else if (consecutiveErrs >= HEARTBEAT_KILL_POLLS) {
             tearing = true;
             stopTicker();
-            onTeardown('heartbeat-lost');
+            await onTeardown('heartbeat-lost');
             return { status: 'failed', exitCode: null, runtimeMs: elapsedMs, failureType: 'infrastructure' };
           }
         }
@@ -202,8 +239,9 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
           // Skip provider util lines — they're shown in the status bar instead.
           if (/\b(gpu_util|cpu_util|provider_status|uptime)=/.test(line)) continue;
 
+          const isErrorLine = /^error\b/i.test(line) || /Error response from daemon/i.test(line);
           stopTicker();
-          console.log(`  ${chalk.dim(line)}`);
+          console.log(`  ${isErrorLine ? chalk.red(line) : chalk.dim(line)}`);
           startTicker();
         }
       } catch {
@@ -347,6 +385,7 @@ export async function runCommand(config, args, chalk) {
       ...(effectiveRegion ? { region: effectiveRegion } : {}),
       max_price_per_hour: flags.maxPrice,
       name: flags.name,
+      ...(flags.tier ? { tier: flags.tier } : {}),
     };
   }
 
@@ -399,7 +438,7 @@ export async function runCommand(config, args, chalk) {
         const d2 = err2.errorData;
         if (d2?.code === 'PROVISIONING_FAILED' || d2?.code === 'PROVIDER_ADAPTER_ERROR') {
           if (d2?.low_cost_provider_failed) {
-            console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+            console.error(chalk.red(`\n  ✗ Tier 2 unavailable. Tier 1 also unavailable. Try again shortly.\n`));
           } else {
             console.error(chalk.red(`\n  ✗ Badgr found ${chosen.gpu} capacity but could not start the machine. Please try again.\n`));
           }
@@ -416,7 +455,7 @@ export async function runCommand(config, args, chalk) {
       }
     } else if (d?.code === 'PROVISIONING_FAILED' || d?.code === 'PROVIDER_ADAPTER_ERROR') {
       if (d?.low_cost_provider_failed) {
-        console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+        console.error(chalk.red(`\n  ✗ Tier 2 unavailable. Tier 1 also unavailable. Try again shortly.\n`));
       } else {
         console.error(chalk.red(`\n  ✗ Badgr found ${gpu} capacity but could not start the machine. Please try again.\n`));
       }
@@ -434,8 +473,8 @@ export async function runCommand(config, args, chalk) {
     }
   }
 
-  if (dep.provider_fallback_note === 'low_cost_provider_failed') {
-    console.log(chalk.yellow('  ℹ Low-cost provider failed. Using primary provider instead.\n'));
+  if (dep.provider_fallback_note === 'tier2_failed_using_tier1') {
+    console.log(chalk.yellow('  ℹ Tier 2 unavailable. Running on Tier 1 instead.\n'));
   }
 
   const rcptId = dep.receipt_id || generateReceiptId();
@@ -451,6 +490,7 @@ export async function runCommand(config, args, chalk) {
   console.log();
   console.log(chalk.bold(`  Job ID:    ${chalk.cyan(dep.deployment_id)}`));
   console.log(`  ${chalk.bold('GPU:')}      ${dep.gpu_type} × ${dep.gpu_count}`);
+  if (dep.tier) console.log(`  ${chalk.bold('Tier:')}     ${dep.tier}`);
   if (dep.cost_per_hour > 0) console.log(`  ${chalk.bold('Rate:')}     $${dep.cost_per_hour.toFixed(2)}/hr`);
   console.log(`  ${chalk.bold('Receipt:')}  ${chalk.dim(rcptId)}`);
 
@@ -460,6 +500,20 @@ export async function runCommand(config, args, chalk) {
     return;
   }
 
+  // If the backend is still running the smoke check, wait for it to finish.
+  if (dep.status === 'starting') {
+    console.log();
+    dep = await waitForRunning(config, dep.deployment_id, chalk);
+  }
+
+  if (dep.status === 'failed') {
+    console.error(chalk.red('\n  ✗ Container failed to start (infrastructure error).\n'));
+    console.error(chalk.dim('  The backend retried automatically. All attempts failed.'));
+    console.error(chalk.dim(`  Contact support with receipt ID: ${rcptId}`));
+    console.log();
+    process.exit(1);
+  }
+
   const ratePerHour = dep.cost_per_hour || 0;
   console.log(chalk.dim('\n  ── Live status (Ctrl+C to stop) ─────────────────────────────────\n'));
 
@@ -523,7 +577,7 @@ export async function runCommand(config, args, chalk) {
     }
     console.log();
     process.exit(exitCode ?? 1);
-  } else {
+  } else if (finalStatus === 'completed' && (exitCode === 0 || exitCode === null)) {
     console.log(chalk.green(`\n  ✓ Complete\n`));
   }
 }

package/src/commands/serve.js

@@ -17,6 +17,7 @@ export function parseServeArgs(args) {
     if (args[i] === '--gpu')        { flags.gpu = args[++i]; i++; continue; }
     if (args[i] === '--count')      { flags.count = parseInt(args[++i], 10); i++; continue; }
     if (args[i] === '--region')     { flags.region = args[++i]; i++; continue; }
+    if (args[i] === '--tier')       { flags.tier = args[++i]; i++; continue; }
     if (args[i] === '--max-price')  { flags.maxPrice = parseFloat(args[++i]); i++; continue; }
     if (args[i] === '--name')       { flags.name = args[++i]; i++; continue; }
     if (args[i] === '--no-wait')    { flags.noWait = true; i++; continue; }
@@ -79,6 +80,7 @@ export async function serveCommand(config, args, chalk) {
       ...(regionOverride || effectiveRegion ? { region: regionOverride || effectiveRegion } : {}),
       max_price_per_hour: flags.maxPrice,
       name: flags.name,
+      ...(flags.tier ? { tier: flags.tier } : {}),
     };
   }
 
@@ -125,7 +127,7 @@ export async function serveCommand(config, args, chalk) {
         const d2 = err2.errorData;
         if (d2?.code === 'PROVISIONING_FAILED' || d2?.code === 'PROVIDER_ADAPTER_ERROR') {
           if (d2?.low_cost_provider_failed) {
-            console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+            console.error(chalk.red(`\n  ✗ Tier 2 unavailable. Tier 1 also unavailable. Try again shortly.\n`));
           } else {
             console.error(chalk.red(`\n  ✗ Badgr found ${chosen.gpu} capacity but could not start the endpoint. Please try again.\n`));
           }
@@ -137,7 +139,7 @@ export async function serveCommand(config, args, chalk) {
       }
     } else if (d?.code === 'PROVISIONING_FAILED' || d?.code === 'PROVIDER_ADAPTER_ERROR') {
       if (d?.low_cost_provider_failed) {
-        console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+        console.error(chalk.red(`\n  ✗ Tier 2 unavailable. Tier 1 also unavailable. Try again shortly.\n`));
       } else {
         console.error(chalk.red(`\n  ✗ Badgr found capacity but could not start the endpoint. Please try again.\n`));
       }
@@ -161,8 +163,8 @@ export async function serveCommand(config, args, chalk) {
     }
   }
 
-  if (dep.provider_fallback_note === 'low_cost_provider_failed') {
-    console.log(chalk.yellow('  ℹ Low-cost provider failed. Using primary provider instead.\n'));
+  if (dep.provider_fallback_note === 'tier2_failed_using_tier1') {
+    console.log(chalk.yellow('  ℹ Tier 2 unavailable. Running on Tier 1 instead.\n'));
   }
 
   addDeployment({
@@ -215,6 +217,7 @@ export async function serveCommand(config, args, chalk) {
   console.log(`  ${chalk.bold('Base URL:')}   ${chalk.cyan(endpointUrl)}`);
   console.log(`  ${chalk.bold('Model:')}      ${dep.model || model}`);
   console.log(`  ${chalk.bold('GPU:')}        ${dep.gpu_type} × ${dep.gpu_count}`);
+  if (dep.tier) console.log(`  ${chalk.bold('Tier:')}       ${dep.tier}`);
   if (dep.cost_per_hour > 0) console.log(`  ${chalk.bold('Rate:')}       $${dep.cost_per_hour.toFixed(2)}/hr`);
   console.log(`  ${chalk.bold('Logs:')}       ${chalk.dim(`badgr logs ${dep.deployment_id}`)}`);
   console.log(`  ${chalk.bold('Stop billing:')} ${chalk.dim(`badgr down ${dep.deployment_id}`)}`);