badgr-cli 1.0.16 → 1.0.18

package/HOW_IT_WORKS.md

@@ -11,7 +11,7 @@ npm install -g badgr-cli
 badgr login
 ```
 
-`badgr login` prompts for your API key and base URL, then writes them to `~/.gpu/config.json`. Every subsequent command reads that file — no env vars required.
+`badgr login` prompts for your API key and base URL, then writes them to `~/.badgr/config.json`. Every subsequent command reads that file — no env vars required.
 
 ---
 
@@ -142,7 +142,7 @@ badgr receipts [n]        # default: last 10
 
 Shows two sets of receipts:
 
-1. **CLI action receipts** — every `badgr serve` / `badgr down` recorded locally in `~/.gpu/deployments.json`, with provider, retries, latency, and cost.
+1. **CLI action receipts** — every `badgr serve` / `badgr down` recorded locally in `~/.badgr/deployments.json`, with provider, retries, latency, and cost.
 2. **Inference receipts** — per-request records fetched from `GET /v1/receipts` on the API (requires `badgr login`).
 
 Every action — including failures — generates a receipt. Receipt IDs are printed on every command output so you can look them up later.
@@ -202,10 +202,10 @@ GPU type aliases are normalized automatically: `rtx-4090`, `rtx4090`, `4090`, `R
 
 ## Local State
 
-All CLI state lives in `~/.gpu/`:
+All CLI state lives in `~/.badgr/`:
 
 ```
-~/.gpu/
+~/.badgr/
   config.json          API key + base URL
   deployments.json     Active deployments + receipt log (last 200)
 ```

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "badgr-cli",
-  "version": "1.0.16",
+  "version": "1.0.18",
   "description": "Badgr, run or serve GPU workloads from one command",
   "type": "module",
   "bin": {

package/src/commands/run.js

@@ -46,18 +46,75 @@ export function classifyFailure(finalStatus, exitCode) {
   return null;
 }
 
+// Lines the log stream never needs to print — we surface them in the status bar instead.
+const LOG_META_RE = /^\[dep-[^\]]+\] (status|gpu|region|endpoint|cost|receipt|provider_status|uptime)=/;
+
+// Extract structured values from provider status lines so we can show them nicely.
+function parseProviderLine(line) {
+  const gpuUtil  = line.match(/\bgpu_util=([\d.]+)%/);
+  const cpuUtil  = line.match(/\bcpu_util=([\d.]+)%/);
+  const ssh      = line.match(/\bssh=(\S+)/);
+  const provSt   = line.match(/\bprovider_status=(\S+)/);
+  return {
+    gpuUtil:  gpuUtil  ? parseFloat(gpuUtil[1])  : null,
+    cpuUtil:  cpuUtil  ? parseFloat(cpuUtil[1])  : null,
+    ssh:      ssh      ? ssh[1]                  : null,
+    providerStatus: provSt ? provSt[1]           : null,
+  };
+}
+
+function renderStatusBar(chalk, { elapsedMs, ratePerHour, gpuUtil, cpuUtil, maxRuntimeMs, maxCost }) {
+  const spent = ratePerHour * (elapsedMs / 3_600_000);
+  const parts = [`⏱ ${fmtRuntime(elapsedMs)}`];
+  if (ratePerHour > 0) parts.push(`$${spent.toFixed(4)} spent`);
+  if (gpuUtil !== null) parts.push(`GPU ${gpuUtil.toFixed(0)}%`);
+  if (cpuUtil !== null) parts.push(`CPU ${cpuUtil.toFixed(0)}%`);
+  if (maxRuntimeMs) {
+    const left = Math.max(0, maxRuntimeMs - elapsedMs);
+    parts.push(`${fmtRuntime(left)} left`);
+  }
+  if (maxCost && ratePerHour > 0) {
+    const budgetLeft = Math.max(0, maxCost - spent);
+    parts.push(`$${budgetLeft.toFixed(4)} budget left`);
+  }
+  parts.push('Ctrl+C to stop');
+  return chalk.dim('  ' + parts.join('  •  '));
+}
+
 async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost = null, ratePerHour = 0, onTeardown }) {
   const TERMINAL = new Set(['stopped', 'failed', 'completed']);
   const POLL_MS  = 4000;
-  let seenLines       = 0;
+  let seenContent     = new Set();  // track by content, not index, to avoid reprinting stale lines
   let lastStatus      = '';
   let consecutiveErrs = 0;
+  let gpuUtil         = null;
+  let cpuUtil         = null;
+  let sshShown        = false;
+  let statusBarActive = false;
   const startMs       = Date.now();
 
+  // Ticker: update status bar in place every second between polls.
+  let tickerInterval = null;
+  const startTicker = () => {
+    if (tickerInterval) return;
+    statusBarActive = true;
+    tickerInterval = setInterval(() => {
+      const bar = renderStatusBar(chalk, {
+        elapsedMs: Date.now() - startMs, ratePerHour, gpuUtil, cpuUtil, maxRuntimeMs, maxCost,
+      });
+      process.stdout.write(`\r${bar}   `);
+    }, 1000);
+  };
+  const stopTicker = () => {
+    if (tickerInterval) { clearInterval(tickerInterval); tickerInterval = null; }
+    if (statusBarActive) { process.stdout.write('\r\x1b[2K'); statusBarActive = false; } // clear line
+  };
+
   let tearing = false;
   const sigintHandler = () => {
     if (tearing) return;
     tearing = true;
+    stopTicker();
     onTeardown('interrupted');
   };
   process.once('SIGINT', sigintHandler);
@@ -71,12 +128,14 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
 
       if (maxCost !== null && spentSoFar >= maxCost) {
         tearing = true;
+        stopTicker();
         onTeardown('max-cost');
         return { status: 'timeout', exitCode: null, runtimeMs: elapsedMs, failureType: null };
       }
 
       if (maxRuntimeMs !== null && elapsedMs >= maxRuntimeMs) {
         tearing = true;
+        stopTicker();
         onTeardown('max-runtime');
         return { status: 'timeout', exitCode: null, runtimeMs: elapsedMs, failureType: null };
       }
@@ -93,9 +152,12 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
         if (lastStatus === 'running') {
           const lostSec = Math.round((consecutiveErrs * POLL_MS) / 1000);
           if (consecutiveErrs === HEARTBEAT_WARN_POLLS) {
+            stopTicker();
             console.log(chalk.yellow(`\n  ⚠ No heartbeat for ${lostSec}s — cloud machine may be unresponsive`));
+            startTicker();
           } else if (consecutiveErrs >= HEARTBEAT_KILL_POLLS) {
             tearing = true;
+            stopTicker();
             onTeardown('heartbeat-lost');
             return { status: 'failed', exitCode: null, runtimeMs: elapsedMs, failureType: 'infrastructure' };
           }
@@ -106,6 +168,7 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
       const status = dep.status;
       if (status !== lastStatus) {
         if (status === 'running' && lastStatus === 'provisioning') {
+          stopTicker();
           console.log(chalk.dim('  [running]'));
         }
         lastStatus = status;
@@ -117,15 +180,41 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
           baseUrl: config.baseUrl,
         });
         const lines = logData?.logs ?? [];
-        for (let i = seenLines; i < lines.length; i++) {
-          console.log(`  ${chalk.dim(lines[i])}`);
+
+        for (const line of lines) {
+          // Extract structured provider values (gpu_util, ssh, etc.) from any line.
+          const parsed = parseProviderLine(line);
+          if (parsed.gpuUtil  !== null) gpuUtil = parsed.gpuUtil;
+          if (parsed.cpuUtil  !== null) cpuUtil = parsed.cpuUtil;
+
+          // Show SSH address once, prominently.
+          if (parsed.ssh && !sshShown) {
+            stopTicker();
+            console.log(`  ${chalk.bold('SSH:')}      ${chalk.cyan(parsed.ssh)}`);
+            sshShown = true;
+            startTicker();
+          }
+
+          // Skip lines we've already printed and pure metadata lines.
+          if (seenContent.has(line)) continue;
+          seenContent.add(line);
+          if (LOG_META_RE.test(line)) continue;
+          // Skip provider util lines — they're shown in the status bar instead.
+          if (/\b(gpu_util|cpu_util|provider_status|uptime)=/.test(line)) continue;
+
+          stopTicker();
+          console.log(`  ${chalk.dim(line)}`);
+          startTicker();
         }
-        seenLines = lines.length;
       } catch {
         // logs not ready yet
       }
 
+      // Start the ticker once the job is confirmed running.
+      if (status === 'running') startTicker();
+
       if (TERMINAL.has(status)) {
+        stopTicker();
         process.removeListener('SIGINT', sigintHandler);
         const exitCode = dep.exit_code ?? null;
         return {
@@ -137,6 +226,7 @@ async function attachToJob(config, depId, { chalk, maxRuntimeMs = null, maxCost
       }
     }
   } finally {
+    stopTicker();
     process.removeListener('SIGINT', sigintHandler);
   }
 }
@@ -308,7 +398,11 @@ export async function runCommand(config, args, chalk) {
       } catch (err2) {
         const d2 = err2.errorData;
         if (d2?.code === 'PROVISIONING_FAILED' || d2?.code === 'PROVIDER_ADAPTER_ERROR') {
-          console.error(chalk.red(`\n  ✗ Badgr found ${chosen.gpu} capacity but could not start the machine. Please try again.\n`));
+          if (d2?.low_cost_provider_failed) {
+            console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+          } else {
+            console.error(chalk.red(`\n  ✗ Badgr found ${chosen.gpu} capacity but could not start the machine. Please try again.\n`));
+          }
           if (process.env.BADGR_DEBUG === '1' || process.env.BADGR_DEBUG === 'true') {
             if (d2?.debug_error) console.error(chalk.dim(`  Provider detail: ${d2.debug_error}`));
           }
@@ -321,7 +415,11 @@ export async function runCommand(config, args, chalk) {
         process.exit(1);
       }
     } else if (d?.code === 'PROVISIONING_FAILED' || d?.code === 'PROVIDER_ADAPTER_ERROR') {
-      console.error(chalk.red(`\n  ✗ Badgr found ${gpu} capacity but could not start the machine. Please try again.\n`));
+      if (d?.low_cost_provider_failed) {
+        console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+      } else {
+        console.error(chalk.red(`\n  ✗ Badgr found ${gpu} capacity but could not start the machine. Please try again.\n`));
+      }
       if (process.env.BADGR_DEBUG === '1' || process.env.BADGR_DEBUG === 'true') {
         if (d?.debug_error) console.error(chalk.dim(`  Provider detail: ${d.debug_error}`));
       } else {
@@ -336,6 +434,10 @@ 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'));
+  }
+
   const rcptId = dep.receipt_id || generateReceiptId();
   addReceipt({
     receiptId:    rcptId,
@@ -359,7 +461,7 @@ export async function runCommand(config, args, chalk) {
   }
 
   const ratePerHour = dep.cost_per_hour || 0;
-  console.log(chalk.dim('\n  ── Streaming logs (Ctrl+C to stop job) ─────────────────────────\n'));
+  console.log(chalk.dim('\n  ── Live status (Ctrl+C to stop) ─────────────────────────────────\n'));
 
   async function teardown(reason) {
     const labels = {

package/src/commands/serve.js

@@ -124,7 +124,11 @@ export async function serveCommand(config, args, chalk) {
       } catch (err2) {
         const d2 = err2.errorData;
         if (d2?.code === 'PROVISIONING_FAILED' || d2?.code === 'PROVIDER_ADAPTER_ERROR') {
-          console.error(chalk.red(`\n  ✗ Badgr found ${chosen.gpu} capacity but could not start the endpoint. Please try again.\n`));
+          if (d2?.low_cost_provider_failed) {
+            console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+          } else {
+            console.error(chalk.red(`\n  ✗ Badgr found ${chosen.gpu} capacity but could not start the endpoint. Please try again.\n`));
+          }
         } else {
           console.error(chalk.red(`\n  ✗ Failed to start endpoint on ${chosen.gpu}: ${err2.message}\n`));
         }
@@ -132,7 +136,11 @@ export async function serveCommand(config, args, chalk) {
         process.exit(1);
       }
     } else if (d?.code === 'PROVISIONING_FAILED' || d?.code === 'PROVIDER_ADAPTER_ERROR') {
-      console.error(chalk.red(`\n  ✗ Badgr found capacity but could not start the endpoint. Please try again.\n`));
+      if (d?.low_cost_provider_failed) {
+        console.error(chalk.red(`\n  ✗ Low-cost provider failed. Primary provider also unavailable.\n`));
+      } else {
+        console.error(chalk.red(`\n  ✗ Badgr found capacity but could not start the endpoint. Please try again.\n`));
+      }
       console.error(chalk.dim(`  Run BADGR_DEBUG=1 badgr serve … for a full trace.\n`));
       process.exit(1);
     } else {
@@ -153,6 +161,10 @@ 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'));
+  }
+
   addDeployment({
     id:          dep.deployment_id,
     name:        dep.name,

package/src/store.js

@@ -1,5 +1,5 @@
 /**
- * Local deployment state — persisted to ~/.gpu/deployments.json.
+ * Local deployment state — persisted to ~/.badgr/deployments.json.
  *
  * Tracks what `gpu up` has provisioned so `gpu down/status/logs/receipts`
  * have something to work with before a backend deployments API exists.