Grant
1889ab45fb
Hotfix (was v0.3.1):
- services.py: cache 'unreachable' per (host,user) for 25s so a dead Spark doesn't hang every /api/services call behind 6s ssh timeout
- ssh_run timeout reduced 10 -> 6s for docker_state probes
- hardware probe: shorter SSH timeout (6s), longer cache TTL for failures (25s)
- JS pollStatus retries loadModels() if state.models is empty (recovers from cold-start proxy timeout)
- Unreachable hardware card now includes troubleshooting steps (Spark Control cannot SSH into an unreachable Spark to restart it)
v0.4 NIM installer:
- nim.py module: curated SUGGESTED_NIMS list (Parakeet, Magpie, Riva) + NimManager that runs docker login nvcr.io + docker pull + docker run -d --gpus all -p PORT:PORT -v VOL:/opt/nim/.cache -e NGC_API_KEY -e ... --restart=unless-stopped + chown the volume to uid 1000 + restart. Streams all output via SSE; redacts the API key from log lines.
- custom_services.py: persists installed NIMs to /data/services-overrides.yaml so they appear in the services panel after install
- services.py: merges custom services into the panel
- /api/nim/catalog GET, /api/nim/install POST + GET/SSE
- /api/services/{name} DELETE for custom services
- UI: '+ Install NIM' button next to 'Always-on services'; modal lists curated images each with a 'Pick' button + a custom-image form; installation runs in a second dialog with phase + elapsed timer + collapsible log
- NGC API key field added to Configure Sparks (masked); injected as NGC_API_KEY env var into the container
Package: bump 0.4.0:0; main.ts adds SERVICES_OVERRIDES + NGC_API_KEY env vars
1332 lines
48 KiB
JavaScript
1332 lines
48 KiB
JavaScript
// spark-control front-end
|
||
|
||
const state = {
|
||
models: {},
|
||
defaults: {},
|
||
current_model_key: null,
|
||
swap_job_id: null,
|
||
swap_eventsource: null,
|
||
swap_started_at: null,
|
||
swap_lines: [], // local accumulator for phase detection
|
||
swap_phase: 'Starting…',
|
||
swap_phase_detail: '',
|
||
swap_progress: 0, // 0–1
|
||
services: {},
|
||
service_action_in_flight: null, // e.g. "parakeet:restart"
|
||
hardware: {},
|
||
config: {},
|
||
configured: true,
|
||
timer_handle: null,
|
||
};
|
||
|
||
const el = (sel) => document.querySelector(sel);
|
||
const $$ = (sel) => document.querySelectorAll(sel);
|
||
|
||
function escapeHtml(s) {
|
||
if (s == null) return '';
|
||
return String(s)
|
||
.replaceAll('&', '&')
|
||
.replaceAll('<', '<')
|
||
.replaceAll('>', '>')
|
||
.replaceAll('"', '"')
|
||
.replaceAll("'", ''');
|
||
}
|
||
|
||
async function fetchJSON(url, opts) {
|
||
const r = await fetch(url, opts);
|
||
if (!r.ok) {
|
||
const text = await r.text().catch(() => '');
|
||
throw new Error(`${r.status} ${r.statusText}: ${text}`);
|
||
}
|
||
return r.json();
|
||
}
|
||
|
||
// ===================== rendering =====================
|
||
|
||
function renderCards() {
|
||
const root = el('#cards');
|
||
root.innerHTML = '';
|
||
const isSwapping = !!state.swap_job_id;
|
||
for (const key of Object.keys(state.models)) {
|
||
const m = state.models[key];
|
||
const isActive = key === state.current_model_key;
|
||
const card = document.createElement('div');
|
||
card.className = 'card' + (isActive ? ' active' : '');
|
||
const desc = m.description
|
||
? `<div class="desc">${escapeHtml(m.description)}</div>`
|
||
: '';
|
||
const customPill = m.custom ? `<span class="tag custom-pill">custom</span>` : '';
|
||
card.innerHTML = `
|
||
<div class="name">${escapeHtml(m.display_name)}</div>
|
||
<div class="meta">
|
||
<span class="tag mode-${m.mode}">${m.mode}</span>
|
||
<span class="tag">${m.size_gb} GB</span>
|
||
${customPill}
|
||
${(m.capabilities || []).map(c => `<span class="tag cap">${escapeHtml(c)}</span>`).join('')}
|
||
</div>
|
||
${desc}
|
||
<div class="muted small repo">
|
||
<a href="https://huggingface.co/${encodeURIComponent(m.repo)}" target="_blank" rel="noopener" title="View on Hugging Face">${escapeHtml(m.repo)} <span class="hf-icon">↗</span></a>
|
||
</div>
|
||
<div class="spacer"></div>
|
||
<div class="card-actions">
|
||
<button class="btn ${isActive ? '' : 'primary'}" data-swap-key="${key}" ${isActive || isSwapping ? 'disabled' : ''}>
|
||
${isActive ? 'Current' : 'Switch to this'}
|
||
</button>
|
||
<button class="btn adv-btn" data-adv-key="${key}" title="Advanced settings">Advanced</button>
|
||
</div>
|
||
`;
|
||
root.appendChild(card);
|
||
}
|
||
for (const btn of root.querySelectorAll('[data-swap-key]')) {
|
||
btn.addEventListener('click', () => triggerSwap(btn.dataset.swapKey));
|
||
}
|
||
for (const btn of root.querySelectorAll('[data-adv-key]')) {
|
||
btn.addEventListener('click', () => openAdvanced(btn.dataset.advKey));
|
||
}
|
||
}
|
||
|
||
function renderCurrent(status) {
|
||
const c = el('#current');
|
||
if (!status.configured) { c.innerHTML = `<span class="muted">not configured</span>`; return; }
|
||
if (status.current_swap_job) { c.innerHTML = `<span class="muted">swap in progress</span>`; return; }
|
||
const v = status.vllm || {};
|
||
if (!v.ok) { c.innerHTML = `<span class="muted">vLLM unreachable</span>`; return; }
|
||
const m = status.current_model_key ? state.models[status.current_model_key] : null;
|
||
const label = m ? m.display_name : (v.current_model || '(unknown)');
|
||
c.innerHTML = `<strong>${label}</strong>`;
|
||
}
|
||
|
||
// ===================== hardware dashboard =====================
|
||
|
||
function fmtBytes(n) {
|
||
if (!n && n !== 0) return '—';
|
||
const u = ['B', 'KB', 'MB', 'GB', 'TB'];
|
||
let i = 0; let v = n;
|
||
while (v >= 1024 && i < u.length - 1) { v /= 1024; i++; }
|
||
return v < 10 ? `${v.toFixed(1)} ${u[i]}` : `${Math.round(v)} ${u[i]}`;
|
||
}
|
||
function fmtMiB(n) {
|
||
if (!n && n !== 0) return null;
|
||
// n is in MiB; render in GB
|
||
const gb = n / 1024;
|
||
return gb < 10 ? gb.toFixed(1) : Math.round(gb).toString();
|
||
}
|
||
|
||
function bar(usedPct, warn) {
|
||
const pct = Math.max(2, Math.min(100, usedPct));
|
||
return `<div class="bar ${warn ? 'warn' : ''}"><span style="width:${pct}%"></span></div>`;
|
||
}
|
||
|
||
async function pollHardware() {
|
||
try {
|
||
state.hardware = await fetchJSON('/api/hardware');
|
||
renderHardware();
|
||
} catch (e) { console.warn('hardware poll failed', e); }
|
||
}
|
||
|
||
function renderHardware() {
|
||
const panel = el('#hardware-panel');
|
||
const grid = el('#hardware-grid');
|
||
const hw = state.hardware || {};
|
||
const keys = Object.keys(hw).filter(k => hw[k] && (hw[k].configured !== false));
|
||
if (keys.length === 0) { panel.classList.add('hidden'); return; }
|
||
panel.classList.remove('hidden');
|
||
grid.innerHTML = '';
|
||
for (const key of keys) {
|
||
const s = hw[key];
|
||
const card = document.createElement('div');
|
||
if (!s.reachable) {
|
||
card.className = 'hw-card unreachable';
|
||
card.innerHTML = `
|
||
<div class="head">
|
||
<span class="name">${escapeHtml(key)}</span>
|
||
<span class="meta">unreachable</span>
|
||
</div>
|
||
<div class="muted small">${escapeHtml(s.host || '')} — ${escapeHtml(s.error || 'no response')}</div>
|
||
<div class="muted small" style="line-height:1.5">
|
||
Spark Control can't restart a Spark that won't answer SSH. Steps to try:
|
||
<ol style="margin: 6px 0 0 18px; padding: 0;">
|
||
<li>Verify it's powered on (check the front LED).</li>
|
||
<li>Ping it from another LAN device.</li>
|
||
<li>Power-cycle it physically.</li>
|
||
<li>If it boots, this card will go green again automatically.</li>
|
||
</ol>
|
||
</div>
|
||
`;
|
||
grid.appendChild(card);
|
||
continue;
|
||
}
|
||
const ramPct = s.ram_used_bytes && s.ram_total_bytes ? (s.ram_used_bytes / s.ram_total_bytes) * 100 : 0;
|
||
const diskPct = s.disk_used_bytes && s.disk_total_bytes ? (s.disk_used_bytes / s.disk_total_bytes) * 100 : 0;
|
||
const loadPct = (s.load && s.cores) ? Math.min(100, (s.load[0] / s.cores) * 100) : 0;
|
||
// GPU memory: on unified-memory systems (DGX Spark) total is N/A, so use system RAM as the pool.
|
||
const gpuMemTotalMiB = s.gpu_mem_total_mib || (s.gpu_unified_memory ? (s.ram_total_bytes / (1024 * 1024)) : null);
|
||
const gpuMemUsedMiB = s.gpu_mem_used_mib ?? null;
|
||
const gpuMemPct = (gpuMemTotalMiB && gpuMemUsedMiB != null) ? (gpuMemUsedMiB / gpuMemTotalMiB) * 100 : 0;
|
||
const gpuMemNote = s.gpu_unified_memory ? ' <span class="muted">(unified)</span>' : '';
|
||
const gpuExtras = [];
|
||
if (s.gpu_temp_c != null) gpuExtras.push(`${s.gpu_temp_c}°C`);
|
||
if (s.gpu_power_w != null) gpuExtras.push(`${s.gpu_power_w.toFixed(0)}W`);
|
||
const gpuExtrasStr = gpuExtras.length ? ` · ${gpuExtras.join(' · ')}` : '';
|
||
card.className = 'hw-card';
|
||
card.innerHTML = `
|
||
<div class="head">
|
||
<span class="name">${escapeHtml(s.hostname || key)}</span>
|
||
<span class="meta">${escapeHtml(key)} · ${escapeHtml(s.gpu_name || '')} · ${escapeHtml(s.uptime || '')}</span>
|
||
</div>
|
||
<div class="hw-metric">
|
||
<span class="label">CPU</span>
|
||
${bar(loadPct, loadPct > 80)}
|
||
<span class="val">${s.load ? s.load[0].toFixed(2) : '—'} / ${s.cores || '?'} cores</span>
|
||
</div>
|
||
<div class="hw-metric">
|
||
<span class="label">RAM</span>
|
||
${bar(ramPct, ramPct > 85)}
|
||
<span class="val">${fmtBytes(s.ram_used_bytes)} / ${fmtBytes(s.ram_total_bytes)}</span>
|
||
</div>
|
||
<div class="hw-metric">
|
||
<span class="label">GPU mem${gpuMemNote}</span>
|
||
${bar(gpuMemPct, gpuMemPct > 90)}
|
||
<span class="val">${fmtMiB(gpuMemUsedMiB) || '—'} / ${fmtMiB(gpuMemTotalMiB) || '?'} GB</span>
|
||
</div>
|
||
<div class="hw-metric">
|
||
<span class="label">GPU util</span>
|
||
${bar(s.gpu_util_pct || 0, (s.gpu_util_pct || 0) > 90)}
|
||
<span class="val">${s.gpu_util_pct ?? 0}%${gpuExtrasStr}</span>
|
||
</div>
|
||
<div class="hw-metric">
|
||
<span class="label">Disk</span>
|
||
${bar(diskPct, diskPct > 85)}
|
||
<span class="val">${fmtBytes(s.disk_used_bytes)} / ${fmtBytes(s.disk_total_bytes)}</span>
|
||
</div>
|
||
`;
|
||
grid.appendChild(card);
|
||
}
|
||
}
|
||
|
||
// ===================== service classification =====================
|
||
|
||
function classifyService(s) {
|
||
// returns one of: running | unhealthy | missing | unconfigured | starting
|
||
if (!s.host) return 'unconfigured';
|
||
if (s.docker_state === 'missing') return 'missing';
|
||
if (s.docker_state === 'restarting') return 'unhealthy';
|
||
if (s.docker_state === 'exited') return 'unhealthy';
|
||
if (s.docker_state === 'running' && !s.http_ready) return 'starting';
|
||
if (s.docker_state === 'running' && s.http_ready) return 'running';
|
||
return s.docker_state || 'unknown';
|
||
}
|
||
|
||
function statusLabel(cls) {
|
||
return {
|
||
running: 'Healthy',
|
||
unhealthy: 'Unhealthy',
|
||
starting: 'Starting…',
|
||
missing: 'Not installed',
|
||
unconfigured: 'Not configured',
|
||
unknown: 'Unknown',
|
||
}[cls] || cls;
|
||
}
|
||
|
||
async function renderServices() {
|
||
let services = state.services;
|
||
// First render: fetch.
|
||
if (!services || Object.keys(services).length === 0) {
|
||
try {
|
||
services = await fetchJSON('/api/services');
|
||
state.services = services;
|
||
} catch (e) { console.error('services fetch failed', e); return; }
|
||
}
|
||
const panel = el('#services-panel');
|
||
const grid = el('#services-grid');
|
||
const entries = Object.entries(services);
|
||
if (entries.length === 0) { panel.classList.add('hidden'); return; }
|
||
panel.classList.remove('hidden');
|
||
grid.innerHTML = '';
|
||
for (const [name, s] of entries) {
|
||
const cls = classifyService(s);
|
||
const card = document.createElement('div');
|
||
card.className = `service-card ${cls}`;
|
||
const inFlight = state.service_action_in_flight && state.service_action_in_flight.startsWith(name + ':');
|
||
const disable = (action) => {
|
||
// Disable buttons that don't make sense for the current state
|
||
if (inFlight) return true;
|
||
if (cls === 'unconfigured' || cls === 'missing') return true;
|
||
if (action === 'start' && (cls === 'running' || cls === 'starting')) return true;
|
||
if (action === 'stop' && cls !== 'running' && cls !== 'starting' && cls !== 'unhealthy') return true;
|
||
return false;
|
||
};
|
||
const copyIcon = `<svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><rect x="9" y="9" width="13" height="13" rx="2"/><path d="M5 15H4a2 2 0 0 1-2-2V4a2 2 0 0 1 2-2h9a2 2 0 0 1 2 2v1"/></svg>`;
|
||
const hostStr = s.host ? `${s.host}:${s.port}` : '';
|
||
const hostRow = s.host
|
||
? `<div class="row"><span class="k">Host</span><span class="v copyable" data-copy-self title="Click to copy">${escapeHtml(hostStr)}</span><button class="icon-btn" data-copy-text="${escapeHtml(hostStr)}" title="Copy host" aria-label="Copy">${copyIcon}</button></div>`
|
||
: `<div class="row"><span class="k">Host</span><span class="v muted-v">not configured</span></div>`;
|
||
const urlRow = s.base_url
|
||
? `<div class="row"><span class="k">URL</span><span class="v copyable" data-copy-self title="Click to copy">${escapeHtml(s.base_url)}</span><button class="icon-btn" data-copy-text="${escapeHtml(s.base_url)}" title="Copy URL" aria-label="Copy">${copyIcon}</button></div>`
|
||
: '';
|
||
const modelRow = s.model
|
||
? `<div class="row"><span class="k">Model</span><span class="v copyable" data-copy-self title="Click to copy">${escapeHtml(s.model)}</span><button class="icon-btn" data-copy-text="${escapeHtml(s.model)}" title="Copy model" aria-label="Copy">${copyIcon}</button></div>`
|
||
: '';
|
||
const restartsRow = s.restart_count != null && s.restart_count > 1
|
||
? `<div class="row"><span class="k">Restarts</span><span class="v">${s.restart_count}</span></div>`
|
||
: '';
|
||
card.innerHTML = `
|
||
<div class="head">
|
||
<span class="name">${escapeHtml(name)}</span>
|
||
<span class="kind">${escapeHtml(s.kind || '')}</span>
|
||
<span class="status">${statusLabel(cls)}</span>
|
||
</div>
|
||
${hostRow}
|
||
${urlRow}
|
||
${modelRow}
|
||
${restartsRow}
|
||
<div class="service-actions">
|
||
<button class="btn" data-svc-action="${name}:start" ${disable('start') ? 'disabled' : ''}>Start</button>
|
||
<button class="btn" data-svc-action="${name}:restart" ${disable('restart') ? 'disabled' : ''}>Restart</button>
|
||
<button class="btn danger" data-svc-action="${name}:stop" ${disable('stop') ? 'disabled' : ''}>Stop</button>
|
||
</div>
|
||
`;
|
||
grid.appendChild(card);
|
||
}
|
||
for (const btn of grid.querySelectorAll('.btn[data-svc-action]')) {
|
||
btn.addEventListener('click', () => onServiceAction(btn.dataset.svcAction));
|
||
}
|
||
}
|
||
|
||
async function onServiceAction(key) {
|
||
if (state.service_action_in_flight) return;
|
||
const [name, action] = key.split(':');
|
||
state.service_action_in_flight = key;
|
||
renderServices();
|
||
try {
|
||
await fetchJSON(`/api/services/${name}/${action}`, { method: 'POST' });
|
||
} catch (e) {
|
||
alert(`${action} ${name} failed: ${e.message}`);
|
||
} finally {
|
||
state.service_action_in_flight = null;
|
||
// Refresh services state
|
||
try {
|
||
state.services = await fetchJSON('/api/services');
|
||
} catch {}
|
||
renderServices();
|
||
pollStatus();
|
||
}
|
||
}
|
||
|
||
function renderEndpoint(status) {
|
||
const v = status.vllm || {};
|
||
const panel = el('#endpoint-panel');
|
||
const ready = v.ok && v.current_model && v.base_url;
|
||
panel.classList.toggle('hidden', !ready);
|
||
if (!ready) return;
|
||
el('#ep-url').textContent = v.base_url;
|
||
el('#ep-model').textContent = v.current_model;
|
||
const snippet =
|
||
`curl -s ${v.base_url}/chat/completions \\
|
||
-H 'content-type: application/json' \\
|
||
-d '{
|
||
"model": "${v.current_model}",
|
||
"messages": [{"role": "user", "content": "Hello"}]
|
||
}'`;
|
||
el('#ep-curl-snippet').textContent = snippet;
|
||
}
|
||
|
||
async function copyText(text, indicatorEl) {
|
||
try {
|
||
await navigator.clipboard.writeText(text);
|
||
if (indicatorEl) {
|
||
indicatorEl.classList.add('copied');
|
||
setTimeout(() => indicatorEl.classList.remove('copied'), 1200);
|
||
}
|
||
return true;
|
||
} catch {
|
||
// Plain HTTP fallback: select the text so the user can ⌘C
|
||
if (indicatorEl) {
|
||
const range = document.createRange();
|
||
range.selectNode(indicatorEl);
|
||
window.getSelection().removeAllRanges();
|
||
window.getSelection().addRange(range);
|
||
}
|
||
return false;
|
||
}
|
||
}
|
||
|
||
function setupCopyButtons() {
|
||
document.body.addEventListener('click', async (e) => {
|
||
// Inline icon copy with literal text (used for dynamically-rendered service rows)
|
||
const litBtn = e.target.closest('[data-copy-text]');
|
||
if (litBtn) {
|
||
await copyText(litBtn.dataset.copyText, litBtn);
|
||
return;
|
||
}
|
||
// Copy buttons (with svg icon) referenced by data-copy="selector"
|
||
const btn = e.target.closest('[data-copy]');
|
||
if (btn) {
|
||
const target = el(btn.dataset.copy);
|
||
if (target) {
|
||
await copyText(target.textContent, btn);
|
||
target.classList.add('copied');
|
||
setTimeout(() => target.classList.remove('copied'), 1200);
|
||
}
|
||
return;
|
||
}
|
||
// Self-copy: clicking the text itself
|
||
const selfCopy = e.target.closest('[data-copy-self]');
|
||
if (selfCopy) {
|
||
await copyText(selfCopy.textContent, selfCopy);
|
||
}
|
||
});
|
||
}
|
||
|
||
function renderHealth(status) {
|
||
function setDot(id, ok, payload) {
|
||
const item = el(id);
|
||
if (!item) return;
|
||
const dot = item.querySelector('.dot');
|
||
dot.classList.remove('ok', 'bad', 'warn');
|
||
if (ok === true) dot.classList.add('ok');
|
||
else if (ok === false) dot.classList.add('bad');
|
||
else dot.classList.add('warn');
|
||
item.title = JSON.stringify(payload || {}, null, 2);
|
||
}
|
||
setDot('#h-vllm', status.vllm && status.vllm.ok, status.vllm);
|
||
setDot('#h-parakeet', status.parakeet && status.parakeet.ok, status.parakeet);
|
||
setDot('#h-magpie', status.magpie && status.magpie.ok, status.magpie);
|
||
el('#updated').textContent = `updated ${new Date().toLocaleTimeString()}`;
|
||
}
|
||
|
||
function renderBanner(status) {
|
||
el('#setup-banner').classList.toggle('hidden', !!status.configured);
|
||
}
|
||
|
||
function renderSwapPanel() {
|
||
el('#swap-phase').textContent = state.swap_phase;
|
||
el('#swap-phase-detail').textContent = state.swap_phase_detail;
|
||
el('#swap-phase-fill').style.width = `${Math.max(2, Math.round(state.swap_progress * 100))}%`;
|
||
}
|
||
|
||
// ===================== phase detection =====================
|
||
|
||
const PHASE_ORDER = [
|
||
['Stopping current model…', 0.08],
|
||
['Starting new model…', 0.16],
|
||
['Joining Ray cluster…', 0.22],
|
||
['Loading weights…', 0.30],
|
||
['Compiling kernels…', 0.78],
|
||
['Warming up…', 0.88],
|
||
['Starting API server…', 0.94],
|
||
['Ready ✓', 1.00],
|
||
['Failed', 1.00],
|
||
];
|
||
|
||
function phaseProgress(name) {
|
||
const found = PHASE_ORDER.find(([n]) => n === name);
|
||
return found ? found[1] : 0.05;
|
||
}
|
||
|
||
function deriveSwapPhase(serverState, lines) {
|
||
// Default phase from server state
|
||
let phase = ({
|
||
starting: 'Starting…',
|
||
stopping: 'Stopping current model…',
|
||
launching: 'Starting new model…',
|
||
tailing: 'Loading weights…',
|
||
ready: 'Ready ✓',
|
||
failed: 'Failed',
|
||
})[serverState] || 'Working…';
|
||
|
||
let detail = '';
|
||
|
||
// Refine from log content (search recent lines first)
|
||
const tail = lines.slice(-40);
|
||
for (let i = tail.length - 1; i >= 0; i--) {
|
||
const line = tail[i];
|
||
if (line.includes('Application startup complete')) {
|
||
phase = 'Ready ✓';
|
||
break;
|
||
}
|
||
if (line.includes('Started server process')) {
|
||
phase = 'Starting API server…';
|
||
break;
|
||
}
|
||
if (line.includes('Initial profiling/warmup') || line.includes('init engine (profile, create kv cache, warmup model)')) {
|
||
phase = 'Warming up…';
|
||
break;
|
||
}
|
||
if (line.match(/Capturing CUDA graphs|Compiling a graph|torch\.compile took|Graph capturing/)) {
|
||
phase = 'Compiling kernels…';
|
||
break;
|
||
}
|
||
const shard = line.match(/Loading safetensors checkpoint shards:\s+(\d+)%\s+Completed\s+\|\s+(\d+)\/(\d+)/);
|
||
if (shard) {
|
||
phase = 'Loading weights…';
|
||
detail = `${shard[2]} of ${shard[3]} shards (${shard[1]}%)`;
|
||
const innerProgress = parseInt(shard[2], 10) / parseInt(shard[3], 10);
|
||
// Map shard progress 0..1 into the 0.30..0.78 band
|
||
state.swap_progress = 0.30 + (0.78 - 0.30) * innerProgress;
|
||
state.swap_phase = phase;
|
||
state.swap_phase_detail = detail;
|
||
return;
|
||
}
|
||
if (line.includes('Connecting to existing Ray cluster')) {
|
||
phase = 'Joining Ray cluster…';
|
||
break;
|
||
}
|
||
if (line.includes('Resolved architecture') || line.match(/launch-cluster\.sh.*exec vllm serve/)) {
|
||
phase = 'Starting new model…';
|
||
break;
|
||
}
|
||
if (line.match(/launch-cluster\.sh stop/)) {
|
||
phase = 'Stopping current model…';
|
||
break;
|
||
}
|
||
}
|
||
|
||
state.swap_phase = phase;
|
||
state.swap_phase_detail = detail;
|
||
state.swap_progress = phaseProgress(phase);
|
||
}
|
||
|
||
// ===================== timer =====================
|
||
|
||
function startTimer(startedAtMillis) {
|
||
state.swap_started_at = startedAtMillis;
|
||
if (state.timer_handle) clearInterval(state.timer_handle);
|
||
const tick = () => {
|
||
if (!state.swap_started_at) return;
|
||
const sec = Math.max(0, Math.floor((Date.now() - state.swap_started_at) / 1000));
|
||
const m = Math.floor(sec / 60);
|
||
const s = sec % 60;
|
||
el('#swap-elapsed').textContent = `${m}:${s.toString().padStart(2, '0')}`;
|
||
};
|
||
tick();
|
||
state.timer_handle = setInterval(tick, 500);
|
||
}
|
||
|
||
function stopTimer() {
|
||
if (state.timer_handle) { clearInterval(state.timer_handle); state.timer_handle = null; }
|
||
}
|
||
|
||
// ===================== polling + SSE =====================
|
||
|
||
async function pollStatus() {
|
||
try {
|
||
const status = await fetchJSON('/api/status');
|
||
state.current_model_key = status.current_model_key;
|
||
state.configured = status.configured;
|
||
renderBanner(status);
|
||
renderCurrent(status);
|
||
renderEndpoint(status);
|
||
renderHealth(status);
|
||
// If models hasn't loaded yet (init may have hit a transient proxy timeout), retry.
|
||
if (!state.models || Object.keys(state.models).length === 0) {
|
||
try { await loadModels(); } catch {}
|
||
}
|
||
// Refresh services state lazily — every 5s poll triggers this too.
|
||
try {
|
||
state.services = await fetchJSON('/api/services');
|
||
renderServices();
|
||
} catch {}
|
||
if (status.current_swap_job && status.current_swap_job !== state.swap_job_id) {
|
||
attachToSwap(status.current_swap_job, /*needsBackfill=*/true);
|
||
} else if (!status.current_swap_job && state.swap_job_id && !state.swap_eventsource) {
|
||
// Foreign swap ended
|
||
detachSwap();
|
||
}
|
||
renderCards();
|
||
} catch (e) {
|
||
console.error('status poll failed', e);
|
||
}
|
||
}
|
||
|
||
async function loadModels() {
|
||
const data = await fetchJSON('/api/models');
|
||
state.defaults = data.defaults || {};
|
||
state.models = data.models || {};
|
||
}
|
||
|
||
async function triggerSwap(modelKey) {
|
||
if (state.swap_job_id) return;
|
||
try {
|
||
const r = await fetchJSON('/api/swap', {
|
||
method: 'POST',
|
||
headers: { 'content-type': 'application/json' },
|
||
body: JSON.stringify({ model_key: modelKey }),
|
||
});
|
||
attachToSwap(r.job_id, /*needsBackfill=*/false);
|
||
} catch (e) {
|
||
alert('Failed to start swap: ' + e.message);
|
||
}
|
||
}
|
||
|
||
async function attachToSwap(jobId, needsBackfill) {
|
||
if (state.swap_eventsource) {
|
||
state.swap_eventsource.close();
|
||
state.swap_eventsource = null;
|
||
}
|
||
state.swap_job_id = jobId;
|
||
state.swap_lines = [];
|
||
state.swap_phase = 'Starting…';
|
||
state.swap_phase_detail = '';
|
||
state.swap_progress = 0.05;
|
||
el('#swap-log').textContent = '';
|
||
el('#swap-panel').classList.remove('hidden');
|
||
renderSwapPanel();
|
||
|
||
// Backfill (if joining mid-swap) — fetch the snapshot so we have started_at + history
|
||
try {
|
||
const snap = await fetchJSON(`/api/swap/${jobId}`);
|
||
const ts = Date.parse(snap.started_at);
|
||
if (!isNaN(ts)) startTimer(ts);
|
||
state.swap_lines = snap.lines || [];
|
||
for (const line of state.swap_lines) appendLog(line);
|
||
deriveSwapPhase(snap.state, state.swap_lines);
|
||
renderSwapPanel();
|
||
if (snap.returncode !== null && snap.returncode !== undefined) {
|
||
// Already finished — close panel after a beat
|
||
handleSwapDone(snap);
|
||
return;
|
||
}
|
||
} catch (e) {
|
||
if (!needsBackfill) startTimer(Date.now());
|
||
console.warn('backfill failed', e);
|
||
}
|
||
|
||
const es = new EventSource(`/api/swap/${jobId}/stream`);
|
||
state.swap_eventsource = es;
|
||
es.onmessage = (ev) => {
|
||
try {
|
||
const d = JSON.parse(ev.data);
|
||
if (d.line !== undefined) {
|
||
state.swap_lines.push(d.line);
|
||
appendLog(d.line);
|
||
deriveSwapPhase(d.state, state.swap_lines);
|
||
renderSwapPanel();
|
||
} else if (d.state) {
|
||
deriveSwapPhase(d.state, state.swap_lines);
|
||
renderSwapPanel();
|
||
}
|
||
} catch {}
|
||
};
|
||
es.addEventListener('done', async (ev) => {
|
||
let d = {};
|
||
try { d = JSON.parse(ev.data); } catch {}
|
||
handleSwapDone(d);
|
||
});
|
||
es.onerror = () => {
|
||
// Tab backgrounded or network blip — close; status poll will reattach
|
||
es.close();
|
||
state.swap_eventsource = null;
|
||
};
|
||
renderCards();
|
||
}
|
||
|
||
function handleSwapDone(d) {
|
||
if (state.swap_eventsource) { state.swap_eventsource.close(); state.swap_eventsource = null; }
|
||
const finalState = d.state || 'ready';
|
||
state.swap_phase = finalState === 'failed' ? 'Failed' : 'Ready ✓';
|
||
state.swap_phase_detail = d.returncode !== undefined ? `exit code ${d.returncode}` : '';
|
||
state.swap_progress = 1.0;
|
||
renderSwapPanel();
|
||
setTimeout(() => detachSwap(), 4000);
|
||
pollStatus();
|
||
}
|
||
|
||
function detachSwap() {
|
||
state.swap_job_id = null;
|
||
if (state.swap_eventsource) { state.swap_eventsource.close(); state.swap_eventsource = null; }
|
||
stopTimer();
|
||
el('#swap-panel').classList.add('hidden');
|
||
renderCards();
|
||
}
|
||
|
||
function appendLog(line) {
|
||
const log = el('#swap-log');
|
||
log.textContent += line + '\n';
|
||
log.scrollTop = log.scrollHeight;
|
||
}
|
||
|
||
// ===================== model downloads =====================
|
||
|
||
const dlState = {
|
||
job_id: null,
|
||
eventsource: null,
|
||
started_at: null,
|
||
timer_handle: null,
|
||
};
|
||
|
||
function openDownloadForm() {
|
||
el('#download-panel').classList.remove('hidden');
|
||
el('#download-form').classList.remove('hidden');
|
||
el('#download-progress').classList.add('hidden');
|
||
el('#dl-repo').focus();
|
||
updateDlHfLink();
|
||
}
|
||
|
||
function updateDlHfLink() {
|
||
const repo = el('#dl-repo').value.trim();
|
||
const link = el('#dl-hf-link');
|
||
if (repo.includes('/')) {
|
||
link.href = `https://huggingface.co/${encodeURIComponent(repo)}`;
|
||
link.classList.remove('hidden');
|
||
} else {
|
||
link.classList.add('hidden');
|
||
}
|
||
}
|
||
|
||
function closeDownloadPanel() {
|
||
el('#download-panel').classList.add('hidden');
|
||
el('#download-form').classList.remove('hidden');
|
||
el('#download-progress').classList.add('hidden');
|
||
el('#dl-repo').value = '';
|
||
}
|
||
|
||
function dlTimerStart(startedAt) {
|
||
dlState.started_at = startedAt;
|
||
if (dlState.timer_handle) clearInterval(dlState.timer_handle);
|
||
const tick = () => {
|
||
if (!dlState.started_at) return;
|
||
const sec = Math.max(0, Math.floor((Date.now() - dlState.started_at) / 1000));
|
||
const m = Math.floor(sec / 60);
|
||
const s = sec % 60;
|
||
el('#dl-elapsed').textContent = `${m}:${s.toString().padStart(2, '0')}`;
|
||
};
|
||
tick();
|
||
dlState.timer_handle = setInterval(tick, 500);
|
||
}
|
||
|
||
function dlTimerStop() {
|
||
if (dlState.timer_handle) { clearInterval(dlState.timer_handle); dlState.timer_handle = null; }
|
||
}
|
||
|
||
async function startDownload() {
|
||
const repo = el('#dl-repo').value.trim();
|
||
const mode = document.querySelector('input[name="dl-mode"]:checked').value;
|
||
if (!repo || !repo.includes('/')) {
|
||
alert('Enter a HuggingFace repo in the form "org/name", e.g. RedHatAI/Qwen3.6-35B-A3B-NVFP4');
|
||
return;
|
||
}
|
||
dlState.last_repo = repo;
|
||
dlState.last_mode = mode;
|
||
try {
|
||
const r = await fetchJSON('/api/download', {
|
||
method: 'POST',
|
||
headers: { 'content-type': 'application/json' },
|
||
body: JSON.stringify({ repo, mode }),
|
||
});
|
||
attachToDownload(r.job_id);
|
||
} catch (e) {
|
||
alert('Failed to start download: ' + e.message);
|
||
}
|
||
}
|
||
|
||
function renderDownloadProgress(p) {
|
||
el('#dl-phase').textContent = p.phase || 'Working…';
|
||
const statsParts = [];
|
||
if (p.downloaded && p.total) statsParts.push(`${p.downloaded} / ${p.total}`);
|
||
if (p.rate) statsParts.push(p.rate);
|
||
if (p.eta) statsParts.push(`ETA ${p.eta}`);
|
||
el('#dl-stats').textContent = statsParts.join(' · ');
|
||
const pct = Math.max(2, Math.min(100, p.percent || 2));
|
||
el('#dl-progress-fill').style.width = `${pct}%`;
|
||
el('#dl-phase-detail').textContent = p.percent > 0 ? `${p.percent.toFixed(1)}%` : '';
|
||
}
|
||
|
||
function dlAppendLog(line) {
|
||
const log = el('#dl-log');
|
||
log.textContent += line + '\n';
|
||
log.scrollTop = log.scrollHeight;
|
||
}
|
||
|
||
async function attachToDownload(jobId) {
|
||
if (dlState.eventsource) { dlState.eventsource.close(); dlState.eventsource = null; }
|
||
dlState.job_id = jobId;
|
||
el('#download-form').classList.add('hidden');
|
||
el('#download-progress').classList.remove('hidden');
|
||
el('#dl-log').textContent = '';
|
||
el('#dl-title').textContent = 'Downloading…';
|
||
|
||
try {
|
||
const snap = await fetchJSON(`/api/download/${jobId}`);
|
||
dlTimerStart(Date.parse(snap.started_at));
|
||
for (const line of snap.lines || []) dlAppendLog(line);
|
||
renderDownloadProgress(snap.progress);
|
||
if (snap.returncode !== null && snap.returncode !== undefined) {
|
||
handleDownloadDone(snap);
|
||
return;
|
||
}
|
||
} catch (e) {
|
||
console.warn('download backfill failed', e);
|
||
dlTimerStart(Date.now());
|
||
}
|
||
|
||
const es = new EventSource(`/api/download/${jobId}/stream`);
|
||
dlState.eventsource = es;
|
||
es.onmessage = (ev) => {
|
||
try {
|
||
const d = JSON.parse(ev.data);
|
||
if (d.line !== undefined) dlAppendLog(d.line);
|
||
} catch {}
|
||
};
|
||
es.addEventListener('progress', (ev) => {
|
||
try { renderDownloadProgress(JSON.parse(ev.data)); } catch {}
|
||
});
|
||
es.addEventListener('done', (ev) => {
|
||
let d = {};
|
||
try { d = JSON.parse(ev.data); } catch {}
|
||
handleDownloadDone(d);
|
||
});
|
||
es.onerror = () => { es.close(); dlState.eventsource = null; };
|
||
}
|
||
|
||
function handleDownloadDone(d) {
|
||
if (dlState.eventsource) { dlState.eventsource.close(); dlState.eventsource = null; }
|
||
dlTimerStop();
|
||
if (d.state === 'failed') {
|
||
el('#dl-title').textContent = `Failed (rc=${d.returncode})`;
|
||
el('#dl-phase').textContent = 'Failed';
|
||
} else {
|
||
el('#dl-title').textContent = 'Done';
|
||
el('#dl-phase').textContent = 'Done ✓';
|
||
el('#dl-progress-fill').style.width = '100%';
|
||
// Offer to add to catalog
|
||
const repo = dlState.last_repo;
|
||
const mode = dlState.last_mode;
|
||
if (repo) {
|
||
setTimeout(() => openCatalogDialog(repo, mode), 600);
|
||
}
|
||
}
|
||
dlState.job_id = null;
|
||
}
|
||
|
||
// ===================== Advanced / Add to catalog =====================
|
||
|
||
function gpuTotalGB(modelMode) {
|
||
// Solo uses Spark 1's GPU only. Cluster shares across both — but loading is per-Spark.
|
||
const s1 = state.hardware?.spark1;
|
||
const s2 = state.hardware?.spark2;
|
||
const g1 = s1?.gpu_mem_total_mib ? s1.gpu_mem_total_mib / 1024 : null;
|
||
const g2 = s2?.gpu_mem_total_mib ? s2.gpu_mem_total_mib / 1024 : null;
|
||
if (modelMode === 'cluster' && g1 && g2) return Math.min(g1, g2); // bottleneck
|
||
return g1 || g2 || null;
|
||
}
|
||
|
||
function knobContextHint(field, value, mode) {
|
||
if (field === 'gpu_memory_utilization') {
|
||
const gb = gpuTotalGB(mode);
|
||
if (!gb) return '';
|
||
const used = (value * gb).toFixed(0);
|
||
const free = (gb - value * gb).toFixed(0);
|
||
return `~${used} GB allocated to model + KV cache · ~${free} GB left for OS, buffers, other GPU workloads.`;
|
||
}
|
||
if (field === 'max_model_len') {
|
||
if (!value) return '';
|
||
const pages = Math.round(value / 350); // ~350 tokens per page
|
||
const kvBytes = (value * 2 * 4 * 32 * 128); // rough fp16 KV cache size for typical 32-layer model
|
||
return `~${pages.toLocaleString()} pages of text (very rough). Larger context = more GPU memory reserved for KV cache.`;
|
||
}
|
||
if (field === 'fastsafetensors') return value ? 'Faster cold-start weight loading.' : 'Standard safetensors loading.';
|
||
if (field === 'prefix_caching') return value ? 'Reuses GPU state for repeated prompt prefixes (e.g. long system prompts).' : 'Off — every request re-processes the full prompt.';
|
||
if (field === 'kv_cache_dtype') return value === 'fp8' ? 'Halves KV cache memory (fits ~2× more context). Quality cost is usually imperceptible.' : 'Default precision.';
|
||
return '';
|
||
}
|
||
|
||
function ensureKnobHint(rowEl, id) {
|
||
let h = rowEl.querySelector(`.knob-hint[data-for="${id}"]`);
|
||
if (!h) {
|
||
h = document.createElement('div');
|
||
h.className = 'knob-hint muted small';
|
||
h.dataset.for = id;
|
||
rowEl.appendChild(h);
|
||
}
|
||
return h;
|
||
}
|
||
|
||
function openAdvanced(key) {
|
||
const m = state.models[key];
|
||
if (!m) return;
|
||
const dlg = el('#advanced-dialog');
|
||
el('#adv-title').textContent = `Advanced — ${m.display_name}`;
|
||
const k = m.effective_knobs || {};
|
||
el('#adv-mml').value = k.max_model_len ?? '';
|
||
el('#adv-gmu').value = k.gpu_memory_utilization ?? 0.85;
|
||
el('#adv-gmu-out').value = parseFloat(el('#adv-gmu').value).toFixed(2);
|
||
el('#adv-fst').checked = !!k.fastsafetensors;
|
||
el('#adv-pcache').checked = !!k.prefix_caching;
|
||
el('#adv-fp8').checked = k.kv_cache_dtype === 'fp8';
|
||
|
||
// Wire up live knob hints
|
||
const updateHints = () => {
|
||
const mml = parseInt(el('#adv-mml').value, 10);
|
||
const gmu = parseFloat(el('#adv-gmu').value);
|
||
ensureKnobHint(el('#adv-mml').parentElement, 'mml').textContent = knobContextHint('max_model_len', mml, m.mode);
|
||
ensureKnobHint(el('#adv-gmu').parentElement, 'gmu').textContent = knobContextHint('gpu_memory_utilization', gmu, m.mode);
|
||
ensureKnobHint(el('#adv-fst').parentElement, 'fst').textContent = knobContextHint('fastsafetensors', el('#adv-fst').checked, m.mode);
|
||
ensureKnobHint(el('#adv-pcache').parentElement, 'pcache').textContent = knobContextHint('prefix_caching', el('#adv-pcache').checked, m.mode);
|
||
ensureKnobHint(el('#adv-fp8').parentElement, 'fp8').textContent = knobContextHint('kv_cache_dtype', el('#adv-fp8').checked ? 'fp8' : 'auto', m.mode);
|
||
};
|
||
updateHints();
|
||
el('#adv-mml').oninput = updateHints;
|
||
el('#adv-gmu').oninput = (e) => { el('#adv-gmu-out').value = parseFloat(e.target.value).toFixed(2); updateHints(); };
|
||
el('#adv-fst').onchange = updateHints;
|
||
el('#adv-pcache').onchange = updateHints;
|
||
el('#adv-fp8').onchange = updateHints;
|
||
const del = el('#adv-delete');
|
||
del.classList.toggle('hidden', !m.custom);
|
||
del.onclick = async () => {
|
||
if (!confirm(`Delete "${m.display_name}" from the catalog? The model weights on disk are NOT deleted.`)) return;
|
||
try {
|
||
await fetchJSON(`/api/models/${encodeURIComponent(key)}`, { method: 'DELETE' });
|
||
dlg.close();
|
||
await loadModels();
|
||
pollStatus();
|
||
} catch (e) { alert('Delete failed: ' + e.message); }
|
||
};
|
||
const form = el('#advanced-form');
|
||
form.onsubmit = async (e) => {
|
||
e.preventDefault();
|
||
const knobs = {};
|
||
const mml = parseInt(el('#adv-mml').value, 10);
|
||
if (Number.isFinite(mml) && mml > 0) knobs.max_model_len = mml;
|
||
const gmu = parseFloat(el('#adv-gmu').value);
|
||
if (Number.isFinite(gmu)) knobs.gpu_memory_utilization = gmu;
|
||
if (el('#adv-fst').checked) knobs.fastsafetensors = true; else knobs.fastsafetensors = false;
|
||
if (el('#adv-pcache').checked) knobs.prefix_caching = true; else knobs.prefix_caching = false;
|
||
knobs.kv_cache_dtype = el('#adv-fp8').checked ? 'fp8' : 'auto';
|
||
try {
|
||
await fetchJSON(`/api/models/${encodeURIComponent(key)}/knobs`, {
|
||
method: 'PUT',
|
||
headers: { 'content-type': 'application/json' },
|
||
body: JSON.stringify({ knobs }),
|
||
});
|
||
dlg.close();
|
||
await loadModels();
|
||
pollStatus();
|
||
} catch (e) { alert('Save failed: ' + e.message); }
|
||
};
|
||
dlg.showModal();
|
||
}
|
||
|
||
function openCatalogDialog(repo, mode) {
|
||
const dlg = el('#catalog-dialog');
|
||
const key = repo.split('/').pop().toLowerCase().replace(/[^a-z0-9_-]/g, '-');
|
||
el('#cd-key').value = key;
|
||
el('#cd-name').value = repo.split('/').pop();
|
||
el('#cd-repo').value = repo;
|
||
el('#cd-size').value = '';
|
||
el('#cd-mode').value = mode || 'solo';
|
||
el('#cd-desc').value = '';
|
||
el('#cd-mml').value = 32768;
|
||
el('#cd-gmu').value = 0.85;
|
||
el('#cd-gmu-out').value = '0.85';
|
||
el('#cd-fst').checked = true;
|
||
el('#cd-pcache').checked = true;
|
||
el('#cd-fp8').checked = true;
|
||
dlg.showModal();
|
||
}
|
||
|
||
function setupCatalogDialog() {
|
||
el('#cd-cancel').addEventListener('click', () => el('#catalog-dialog').close());
|
||
el('#cd-gmu').addEventListener('input', (e) => { el('#cd-gmu-out').value = parseFloat(e.target.value).toFixed(2); });
|
||
el('#catalog-form').addEventListener('submit', async (e) => {
|
||
e.preventDefault();
|
||
const body = {
|
||
key: el('#cd-key').value.trim(),
|
||
display_name: el('#cd-name').value.trim(),
|
||
repo: el('#cd-repo').value.trim(),
|
||
size_gb: parseFloat(el('#cd-size').value) || 0,
|
||
mode: el('#cd-mode').value,
|
||
description: el('#cd-desc').value.trim() || null,
|
||
vllm_args: [],
|
||
knobs: {
|
||
max_model_len: parseInt(el('#cd-mml').value, 10) || 32768,
|
||
gpu_memory_utilization: parseFloat(el('#cd-gmu').value),
|
||
fastsafetensors: el('#cd-fst').checked,
|
||
prefix_caching: el('#cd-pcache').checked,
|
||
kv_cache_dtype: el('#cd-fp8').checked ? 'fp8' : 'auto',
|
||
},
|
||
};
|
||
try {
|
||
await fetchJSON('/api/models', {
|
||
method: 'POST',
|
||
headers: { 'content-type': 'application/json' },
|
||
body: JSON.stringify(body),
|
||
});
|
||
el('#catalog-dialog').close();
|
||
closeDownloadPanel();
|
||
await loadModels();
|
||
pollStatus();
|
||
} catch (e) { alert('Add to catalog failed: ' + e.message); }
|
||
});
|
||
}
|
||
|
||
function setupAdvancedDialog() {
|
||
el('#adv-cancel').addEventListener('click', () => el('#advanced-dialog').close());
|
||
el('#adv-gmu').addEventListener('input', (e) => { el('#adv-gmu-out').value = parseFloat(e.target.value).toFixed(2); });
|
||
}
|
||
|
||
// ===================== NIM installer =====================
|
||
|
||
const nimState = {
|
||
catalog: null,
|
||
job_id: null,
|
||
eventsource: null,
|
||
timer: null,
|
||
started_at: null,
|
||
};
|
||
|
||
async function loadNimCatalog() {
|
||
try {
|
||
nimState.catalog = await fetchJSON('/api/nim/catalog');
|
||
el('#nim-catalog-link').href = nimState.catalog.catalog_url;
|
||
const warn = el('#nim-key-warn');
|
||
if (!nimState.catalog.ngc_key_configured) {
|
||
warn.classList.add('nim-key-warn');
|
||
warn.innerHTML = '⚠️ NGC API key not set. Open <strong>Configure Sparks</strong> in StartOS and paste your NGC personal API key, otherwise installs will fail. <a href="https://ngc.nvidia.com/setup/personal-key" target="_blank" rel="noopener">Get a key</a>';
|
||
} else {
|
||
warn.classList.remove('nim-key-warn');
|
||
warn.textContent = '';
|
||
}
|
||
const grid = el('#nim-suggested');
|
||
grid.innerHTML = '';
|
||
for (const s of nimState.catalog.suggested || []) {
|
||
const card = document.createElement('div');
|
||
card.className = 'nim-card';
|
||
card.innerHTML = `
|
||
<div class="info">
|
||
<div class="name">${escapeHtml(s.name)} <span class="muted small">· ${escapeHtml(s.kind || 'nim')}</span></div>
|
||
<div class="desc">${escapeHtml(s.description || '')}</div>
|
||
<div class="img">${escapeHtml(s.image)}</div>
|
||
<div class="links">${s.homepage ? `<a href="${escapeHtml(s.homepage)}" target="_blank" rel="noopener">View on NGC ↗</a>` : ''}</div>
|
||
</div>
|
||
<button type="button" class="btn primary nim-pick" data-image="${escapeHtml(s.image)}" data-container="${escapeHtml(s.default_container)}" data-port="${s.default_port}" data-kind="${escapeHtml(s.kind)}">Pick</button>
|
||
`;
|
||
grid.appendChild(card);
|
||
}
|
||
grid.querySelectorAll('.nim-pick').forEach(btn => {
|
||
btn.addEventListener('click', () => {
|
||
el('#nim-image').value = btn.dataset.image;
|
||
el('#nim-container').value = btn.dataset.container;
|
||
el('#nim-port').value = btn.dataset.port;
|
||
el('#nim-kind').value = btn.dataset.kind || 'nim';
|
||
});
|
||
});
|
||
} catch (e) { console.warn('nim catalog failed', e); }
|
||
}
|
||
|
||
function openNimDialog() {
|
||
loadNimCatalog();
|
||
el('#nim-dialog').showModal();
|
||
}
|
||
|
||
async function submitNim(e) {
|
||
e.preventDefault();
|
||
const body = {
|
||
image: el('#nim-image').value.trim(),
|
||
container: el('#nim-container').value.trim(),
|
||
port: parseInt(el('#nim-port').value, 10),
|
||
host: el('#nim-host').value,
|
||
kind: el('#nim-kind').value,
|
||
};
|
||
if (!body.image || !body.container || !body.port) {
|
||
alert('Image, container name, and port are required.');
|
||
return;
|
||
}
|
||
try {
|
||
const r = await fetchJSON('/api/nim/install', {
|
||
method: 'POST',
|
||
headers: { 'content-type': 'application/json' },
|
||
body: JSON.stringify(body),
|
||
});
|
||
el('#nim-dialog').close();
|
||
attachNimProgress(r.job_id);
|
||
} catch (e) {
|
||
alert('Install failed: ' + e.message);
|
||
}
|
||
}
|
||
|
||
function nimTimerStart(at) {
|
||
nimState.started_at = at;
|
||
if (nimState.timer) clearInterval(nimState.timer);
|
||
const tick = () => {
|
||
if (!nimState.started_at) return;
|
||
const sec = Math.max(0, Math.floor((Date.now() - nimState.started_at) / 1000));
|
||
const m = Math.floor(sec / 60);
|
||
const s = sec % 60;
|
||
el('#nim-prog-elapsed').textContent = `${m}:${s.toString().padStart(2, '0')}`;
|
||
};
|
||
tick();
|
||
nimState.timer = setInterval(tick, 500);
|
||
}
|
||
|
||
async function attachNimProgress(jobId) {
|
||
nimState.job_id = jobId;
|
||
el('#nim-prog-log').textContent = '';
|
||
el('#nim-prog-title').textContent = 'Installing…';
|
||
el('#nim-progress-dialog').showModal();
|
||
try {
|
||
const snap = await fetchJSON(`/api/nim/install/${jobId}`);
|
||
nimTimerStart(Date.parse(snap.started_at));
|
||
el('#nim-prog-phase').textContent = snap.phase || 'Working…';
|
||
el('#nim-prog-log').textContent = (snap.lines || []).join('\n');
|
||
if (snap.returncode !== null) { onNimDone(snap); return; }
|
||
} catch { nimTimerStart(Date.now()); }
|
||
const es = new EventSource(`/api/nim/install/${jobId}/stream`);
|
||
nimState.eventsource = es;
|
||
es.onmessage = ev => {
|
||
try {
|
||
const d = JSON.parse(ev.data);
|
||
if (d.line !== undefined) {
|
||
const log = el('#nim-prog-log');
|
||
log.textContent += d.line + '\n';
|
||
log.scrollTop = log.scrollHeight;
|
||
}
|
||
} catch {}
|
||
};
|
||
es.addEventListener('phase', ev => {
|
||
try { el('#nim-prog-phase').textContent = JSON.parse(ev.data).phase; } catch {}
|
||
});
|
||
es.addEventListener('done', ev => {
|
||
let d = {}; try { d = JSON.parse(ev.data); } catch {}
|
||
onNimDone(d);
|
||
});
|
||
es.onerror = () => { es.close(); nimState.eventsource = null; };
|
||
}
|
||
|
||
function onNimDone(d) {
|
||
if (nimState.eventsource) { nimState.eventsource.close(); nimState.eventsource = null; }
|
||
if (nimState.timer) { clearInterval(nimState.timer); nimState.timer = null; }
|
||
if (d.state === 'failed') {
|
||
el('#nim-prog-title').textContent = `Failed (rc=${d.returncode})`;
|
||
el('#nim-prog-phase').textContent = 'Failed';
|
||
} else {
|
||
el('#nim-prog-title').textContent = 'Installed';
|
||
el('#nim-prog-phase').textContent = 'Done ✓ — service will appear when the container reports healthy.';
|
||
}
|
||
pollStatus();
|
||
}
|
||
|
||
// ===================== Explain context (LLM commit summary) =====================
|
||
|
||
let explainEventSource = null;
|
||
|
||
async function explainContext() {
|
||
if (explainEventSource) { explainEventSource.close(); explainEventSource = null; }
|
||
const section = el('#ub-explain-section');
|
||
const content = el('#ub-explain-content');
|
||
section.classList.remove('hidden');
|
||
section.open = true;
|
||
content.innerHTML = '<span class="muted">Asking the loaded model…</span>';
|
||
let text = '';
|
||
const es = new EventSource('/api/explain-updates');
|
||
explainEventSource = es;
|
||
let firstChunk = true;
|
||
es.onmessage = (ev) => {
|
||
try {
|
||
const d = JSON.parse(ev.data);
|
||
if (d.error) {
|
||
content.innerHTML = `<span class="muted">Couldn't get explanation: ${escapeHtml(d.error)}</span>`;
|
||
return;
|
||
}
|
||
if (firstChunk) { content.innerHTML = ''; firstChunk = false; }
|
||
if (d.content) {
|
||
text += d.content;
|
||
content.textContent = text;
|
||
content.scrollTop = content.scrollHeight;
|
||
} else if (d.reasoning) {
|
||
// Show reasoning tokens but de-emphasized
|
||
let r = content.querySelector('.reasoning-current');
|
||
if (!r) {
|
||
r = document.createElement('div');
|
||
r.className = 'reasoning reasoning-current';
|
||
r.textContent = '';
|
||
content.appendChild(r);
|
||
}
|
||
r.textContent += d.reasoning;
|
||
}
|
||
} catch {}
|
||
};
|
||
es.addEventListener('done', () => {
|
||
es.close();
|
||
explainEventSource = null;
|
||
// strip the reasoning-current marker
|
||
const r = content.querySelector('.reasoning-current');
|
||
if (r) r.classList.remove('reasoning-current');
|
||
});
|
||
es.onerror = () => { es.close(); explainEventSource = null; };
|
||
}
|
||
|
||
// ===================== updates (spark-vllm-docker) =====================
|
||
|
||
const updState = {
|
||
info: null,
|
||
job_id: null,
|
||
eventsource: null,
|
||
started_at: null,
|
||
timer_handle: null,
|
||
};
|
||
|
||
async function pollUpdates() {
|
||
try {
|
||
const info = await fetchJSON('/api/updates');
|
||
updState.info = info;
|
||
renderUpdateBanner();
|
||
} catch (e) {
|
||
console.warn('updates poll failed', e);
|
||
}
|
||
}
|
||
|
||
function renderUpdateBanner() {
|
||
const banner = el('#update-banner');
|
||
const info = updState.info;
|
||
const text = el('#ub-text');
|
||
const details = el('#ub-details');
|
||
const apply = el('#ub-apply');
|
||
const list = el('#ub-list');
|
||
const log = el('#ub-log');
|
||
|
||
if (!info || !info.ok) {
|
||
banner.classList.add('hidden');
|
||
return;
|
||
}
|
||
banner.classList.remove('hidden');
|
||
const behind = info.behind || 0;
|
||
const dirty = info.dirty || 0;
|
||
banner.classList.toggle('up-to-date', behind === 0 && !dirty);
|
||
banner.classList.toggle('warn', !!dirty);
|
||
|
||
const explain = el('#ub-explain');
|
||
if (dirty > 0) {
|
||
text.textContent = `${dirty} local change${dirty === 1 ? '' : 's'} in ~/spark-vllm-docker. Resolve before updating.`;
|
||
details.classList.add('hidden');
|
||
apply.classList.add('hidden');
|
||
explain.classList.add('hidden');
|
||
} else if (behind === 0) {
|
||
text.textContent = `spark-vllm-docker is up to date (${info.current || ''})`;
|
||
details.classList.add('hidden');
|
||
apply.classList.add('hidden');
|
||
list.classList.add('hidden');
|
||
explain.classList.add('hidden');
|
||
} else {
|
||
text.textContent = `${behind} commit${behind === 1 ? '' : 's'} behind upstream`;
|
||
details.classList.remove('hidden');
|
||
apply.classList.remove('hidden');
|
||
explain.classList.remove('hidden');
|
||
log.textContent = (info.log || []).join('\n') || '(no log)';
|
||
}
|
||
}
|
||
|
||
function ubTimerStart(startedAt) {
|
||
updState.started_at = startedAt;
|
||
if (updState.timer_handle) clearInterval(updState.timer_handle);
|
||
const tick = () => {
|
||
if (!updState.started_at) return;
|
||
const sec = Math.max(0, Math.floor((Date.now() - updState.started_at) / 1000));
|
||
const m = Math.floor(sec / 60);
|
||
const s = sec % 60;
|
||
el('#ub-elapsed').textContent = `${m}:${s.toString().padStart(2, '0')}`;
|
||
};
|
||
tick();
|
||
updState.timer_handle = setInterval(tick, 500);
|
||
}
|
||
|
||
async function applyUpdate() {
|
||
if (!confirm('This pulls the latest spark-vllm-docker and rebuilds the vLLM container. Can take 5–40 minutes; the cluster is unaffected until you swap to a different model. Continue?')) return;
|
||
try {
|
||
const r = await fetchJSON('/api/updates/apply', {
|
||
method: 'POST',
|
||
headers: { 'content-type': 'application/json' },
|
||
body: JSON.stringify({ mode: 'cluster' }),
|
||
});
|
||
attachToUpdate(r.job_id);
|
||
} catch (e) {
|
||
alert('Failed to start update: ' + e.message);
|
||
}
|
||
}
|
||
|
||
async function attachToUpdate(jobId) {
|
||
updState.job_id = jobId;
|
||
el('#ub-progress').classList.remove('hidden');
|
||
el('#ub-apply').classList.add('hidden');
|
||
el('#ub-stream').textContent = '';
|
||
el('#ub-phase').textContent = 'Starting…';
|
||
try {
|
||
const snap = await fetchJSON(`/api/updates/${jobId}`);
|
||
ubTimerStart(Date.parse(snap.started_at));
|
||
el('#ub-phase').textContent = snap.phase || 'Working…';
|
||
el('#ub-stream').textContent = (snap.lines || []).join('\n');
|
||
if (snap.returncode !== null) { handleUpdateDone(snap); return; }
|
||
} catch (e) {
|
||
ubTimerStart(Date.now());
|
||
}
|
||
const es = new EventSource(`/api/updates/${jobId}/stream`);
|
||
updState.eventsource = es;
|
||
es.onmessage = (ev) => {
|
||
try {
|
||
const d = JSON.parse(ev.data);
|
||
if (d.line !== undefined) {
|
||
const log = el('#ub-stream');
|
||
log.textContent += d.line + '\n';
|
||
log.scrollTop = log.scrollHeight;
|
||
}
|
||
} catch {}
|
||
};
|
||
es.addEventListener('phase', (ev) => {
|
||
try { el('#ub-phase').textContent = JSON.parse(ev.data).phase; } catch {}
|
||
});
|
||
es.addEventListener('done', (ev) => {
|
||
let d = {}; try { d = JSON.parse(ev.data); } catch {}
|
||
handleUpdateDone(d);
|
||
});
|
||
es.onerror = () => { es.close(); updState.eventsource = null; };
|
||
}
|
||
|
||
function handleUpdateDone(d) {
|
||
if (updState.eventsource) { updState.eventsource.close(); updState.eventsource = null; }
|
||
if (updState.timer_handle) { clearInterval(updState.timer_handle); updState.timer_handle = null; }
|
||
el('#ub-phase').textContent = d.state === 'failed' ? `Failed (rc=${d.returncode})` : 'Done ✓ — re-check from the banner.';
|
||
setTimeout(pollUpdates, 2000);
|
||
}
|
||
|
||
async function init() {
|
||
setupCopyButtons();
|
||
el('#open-download').addEventListener('click', openDownloadForm);
|
||
el('#dl-cancel').addEventListener('click', closeDownloadPanel);
|
||
el('#dl-start').addEventListener('click', startDownload);
|
||
el('#dl-repo').addEventListener('keydown', (e) => { if (e.key === 'Enter') startDownload(); });
|
||
el('#ub-details').addEventListener('click', () => {
|
||
const list = el('#ub-list');
|
||
list.classList.toggle('hidden');
|
||
list.open = !list.open;
|
||
});
|
||
el('#ub-apply').addEventListener('click', applyUpdate);
|
||
el('#ub-explain').addEventListener('click', explainContext);
|
||
el('#dl-repo').addEventListener('input', updateDlHfLink);
|
||
el('#open-nim').addEventListener('click', openNimDialog);
|
||
el('#nim-cancel').addEventListener('click', () => el('#nim-dialog').close());
|
||
el('#nim-form').addEventListener('submit', submitNim);
|
||
el('#nim-prog-close').addEventListener('click', () => el('#nim-progress-dialog').close());
|
||
setupCatalogDialog();
|
||
setupAdvancedDialog();
|
||
// Open WebUI link from /api/config
|
||
try {
|
||
state.config = await fetchJSON('/api/config');
|
||
if (state.config.open_webui_url) {
|
||
const a = el('#open-webui-link');
|
||
a.href = state.config.open_webui_url;
|
||
a.classList.remove('hidden');
|
||
}
|
||
} catch {}
|
||
await loadModels();
|
||
await pollStatus();
|
||
await renderServices();
|
||
pollHardware();
|
||
pollUpdates();
|
||
setInterval(pollStatus, 5000);
|
||
setInterval(pollHardware, 8000); // every 8s
|
||
setInterval(pollUpdates, 300000); // every 5 min
|
||
}
|
||
|
||
init();
|