Release v0.3.2
This commit is contained in:
John Haverlack
commit
b206670478
|
|
@ -5,7 +5,7 @@
|
|||
| **Author** | John Haverlack |
|
||||
| **Copyright** | 2026 John Haverlack |
|
||||
| **License** | MIT |
|
||||
| **Version** | 0.3.1 |
|
||||
| **Version** | 0.3.2 |
|
||||
| **Date** | 2026-01-31 |
|
||||
|
||||
## Overview
|
||||
|
|
@ -23,7 +23,7 @@ See also: [Installation](docs/INSTALL.md)
|
|||
```
|
||||
$ git clone https://github.com/jhaverlack/sdl.git
|
||||
$ cd sdl
|
||||
$ install-sdl.sh
|
||||
$ ./install-sdl.sh
|
||||
```
|
||||
|
||||
Point you browser to http://IP-ADDR:8081
|
||||
|
|
|
|||
|
|
@ -1,10 +1,17 @@
|
|||
{
|
||||
"releases": [
|
||||
{
|
||||
"version": "0.3.2",
|
||||
"date": "2026-01-31",
|
||||
"maturity": "ALPHA",
|
||||
"summary": "",
|
||||
"notes": ""
|
||||
},
|
||||
{
|
||||
"version": "0.3.1",
|
||||
"date": "2026-01-31",
|
||||
"maturity": "ALPHA",
|
||||
"summary": "",
|
||||
"summary": "README Cleanup",
|
||||
"notes": ""
|
||||
},
|
||||
{
|
||||
|
|
@ -12,7 +19,7 @@
|
|||
"date": "2026-01-31",
|
||||
"maturity": "ALPHA",
|
||||
"summary": "Initial Codebase",
|
||||
"notes": ""
|
||||
"notes": "Initial GitHub Release"
|
||||
}
|
||||
]
|
||||
}
|
||||
|
|
|
|||
|
|
@ -5,11 +5,26 @@
|
|||
| **Author** | John Haverlack |
|
||||
| **Copyright** | 2026 John Haverlack |
|
||||
| **License** | MIT |
|
||||
| **Version** | 0.3.1 |
|
||||
| **Version** | 0.3.2 |
|
||||
| **Date** | 2026-01-31 |
|
||||
|
||||
## v0.3.2 - 2026-01-31 (ALPHA)
|
||||
|
||||
- v0.3.2
|
||||
- Squetching Logs
|
||||
- WORKING: join/authz
|
||||
- WIP: Join Worker
|
||||
- WIP: Join Process
|
||||
- sdl-wkr Cluster Join
|
||||
- Fixing Service Module Render Bug
|
||||
- Start v0.3.2
|
||||
|
||||
## v0.3.1 - 2026-01-31 (ALPHA)
|
||||
|
||||
**Summary**
|
||||
README Cleanup
|
||||
|
||||
- v0.3.1
|
||||
- Start v0.3.1
|
||||
- Removing duplicate LICENSE file
|
||||
- README Cleanup
|
||||
|
|
@ -19,6 +34,9 @@
|
|||
**Summary**
|
||||
Initial Codebase
|
||||
|
||||
**Notes**
|
||||
Initial GitHub Release
|
||||
|
||||
- Preparing GitHub Version
|
||||
- Start v0.3.0
|
||||
- Initial commit
|
||||
|
|
|
|||
|
|
@ -5,9 +5,12 @@
|
|||
| **Author** | John Haverlack |
|
||||
| **Copyright** | 2026 John Haverlack |
|
||||
| **License** | MIT |
|
||||
| **Version** | 0.3.1 |
|
||||
| **Version** | 0.3.2 |
|
||||
| **Date** | 2026-01-31 |
|
||||
|
||||
## Overview
|
||||
|
||||
The Software Defined Laboratory (SDL) project provides a generalized distributed computing platform for managing parallel computational workflows across a cluster of distributed nodes. SDL is a minimalistic High Performance Compute (HPC) platform focused on extreme minimizaton of technical debt associated configuration, deployability, and maintenance of the cluster.
|
||||
|
||||
# Roadmap
|
||||
|
||||
|
|
@ -26,16 +29,16 @@ MVP - Minimum Viable Product Task List
|
|||
- [x] SDL_ID Node ID
|
||||
- [x] Version Updater Script
|
||||
- [x] Deploy sld-mgr to $SDL_HOME/sdl-mgr
|
||||
- [ ] Create Worker Join Process
|
||||
- [ ] Create Worker Heartbeat Process
|
||||
- [x] Create SDL Worker Dist Build Process
|
||||
- [x] Create SDL Worker Install to $SDL_HOME
|
||||
- [x] SDL Install Script
|
||||
- [ ] Create SDL Worker Auto Update Process
|
||||
- [x] Create Worker Join Process
|
||||
- [ ] Create Worker Telementry Process
|
||||
- [ ] Data Storage Organizational Structure
|
||||
- [ ] MinIO S3 Storage Server
|
||||
- [ ] Worker Heartbeat Hardware Inventory (CPU, RAM, GPU, etc)
|
||||
- [ ] Create SDL Worker Auto Update Process
|
||||
- [ ] Worker Heartbeat Load Status (CPU, RAM, GPU, Disk IO, etc)
|
||||
- [ ] Worker Telemetry Hardware Inventory (CPU, RAM, GPU, etc)
|
||||
- [ ] Worker Telementry Load Status (CPU, RAM, GPU, Disk IO, etc)
|
||||
|
||||
# Design
|
||||
|
||||
|
|
|
|||
|
|
@ -49,7 +49,7 @@ git pull origin main
|
|||
(Debian, Ubuntu, Linux Mint, Zorin OS, Raspbian)
|
||||
|
||||
```
|
||||
sudo apt -y install gawk coreutils curl jq grep sed sha256sum tar unzip netcat-openbsd
|
||||
sudo apt -y install gawk coreutils curl jq grep sed tar unzip netcat-openbsd
|
||||
```
|
||||
|
||||
#### Installation
|
||||
|
|
|
|||
|
|
@ -3,7 +3,7 @@
|
|||
"name": "sdl",
|
||||
"description": "Software Defined Laboratory",
|
||||
"abbr": "SDL",
|
||||
"version": "0.3.1",
|
||||
"version": "0.3.2",
|
||||
"version_date": "2026-01-31",
|
||||
"maturity": "ALPHA",
|
||||
"author": "John Haverlack",
|
||||
|
|
|
|||
|
|
@ -11,7 +11,7 @@ const config = load_config();
|
|||
log('=======================================================================================');
|
||||
log(config.package.name + ': STARTING: ' + config.package.description + ' v' + config.package.version);
|
||||
// log(JSON.stringify(config, null, 2));
|
||||
log(JSON.stringify(config.dirs, null, 2), false);
|
||||
// log(JSON.stringify(config.dirs, null, 2), false);
|
||||
|
||||
for (let m in config.modules) {
|
||||
// log(`Loading module: ${m}`);
|
||||
|
|
|
|||
|
|
@ -88,6 +88,6 @@ broker.on('publish', (packet, client) => {
|
|||
// console.log(
|
||||
// `Publish from ${client.id}: ${packet.topic}`
|
||||
// );
|
||||
log(`${module}: Publish from ${client.id}: ${packet.topic}`);
|
||||
// log(`${module}: Publish from ${client.id}: ${packet.topic}`);
|
||||
}
|
||||
});
|
||||
|
|
|
|||
|
|
@ -146,6 +146,9 @@ function load_config() {
|
|||
config.package = JSON.parse(fs.readFileSync(path.join(config.dirs.app, 'package.json'), 'utf8'));
|
||||
config.dependencies = {}
|
||||
|
||||
// SDL Version
|
||||
config.cluster.sdl_version = config.package.version
|
||||
|
||||
for (let d in config.package.dependencies) {
|
||||
config.dependencies[d] = JSON.parse(fs.readFileSync(path.join(config.dirs.app, 'node_modules', d, 'package.json'), 'utf8'));
|
||||
}
|
||||
|
|
@ -158,6 +161,7 @@ function load_config() {
|
|||
ips: [],
|
||||
|
||||
os: {
|
||||
platform: null,
|
||||
id: null,
|
||||
name: null,
|
||||
version: null,
|
||||
|
|
@ -181,6 +185,10 @@ function load_config() {
|
|||
|
||||
};
|
||||
|
||||
config.sdl = {
|
||||
version: config.package.version
|
||||
}
|
||||
|
||||
config.identity = {
|
||||
sdl_id: null,
|
||||
created_at: null,
|
||||
|
|
@ -218,6 +226,8 @@ function load_config() {
|
|||
config.host.cpu.cores_logical = cpus.length;
|
||||
}
|
||||
|
||||
config.host.os.platform = os.platform();
|
||||
|
||||
// -------------------------------
|
||||
// OS Info from /etc/os-release (Linux)
|
||||
// -------------------------------
|
||||
|
|
|
|||
|
|
@ -1,14 +1,304 @@
|
|||
const module = 'sdl-mgr'; // Module Name
|
||||
|
||||
import { load_config, log, ipToInt, intToIp, computeBroadcast, getUptimeDHMS } from '../nwa-lib/index.js';
|
||||
import { load_config, log, ipToInt, intToIp, computeBroadcast, getProcessStartTs, getUptimeDHMS } from '../nwa-lib/index.js';
|
||||
import mqtt from 'mqtt';
|
||||
import os from 'os';
|
||||
import dgram from 'dgram';
|
||||
import fs from 'fs';
|
||||
import path from 'path';
|
||||
|
||||
|
||||
|
||||
const config = load_config();
|
||||
|
||||
log(`Loaded module: ${module}`);
|
||||
// log(`${module}: Cluster Conf: ${JSON.stringify(config.cluster, null, 2)}`, true);
|
||||
// log(`${module}: Dirs: ${JSON.stringify(config.dirs, null, 2)}`, true);
|
||||
|
||||
function loadClusterState(config) {
|
||||
const clusterDir = config.dirs.clstr;
|
||||
const clusterFile = path.join(clusterDir, 'cluster.json');
|
||||
|
||||
// Ensure cluster directory exists
|
||||
if (!fs.existsSync(clusterDir)) {
|
||||
fs.mkdirSync(clusterDir, { recursive: true });
|
||||
}
|
||||
|
||||
// If cluster.json doesn't exist, create initial state
|
||||
if (!fs.existsSync(clusterFile)) {
|
||||
const initialState = {
|
||||
meta: {
|
||||
updated: new Date().toISOString(),
|
||||
started_at: new Date(getProcessStartTs()).toISOString(),
|
||||
uptime: getUptimeDHMS(),
|
||||
stats: {
|
||||
workers: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
},
|
||||
resources: {
|
||||
cpus: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
},
|
||||
memory: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
},
|
||||
gpus: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
cluster: config.cluster,
|
||||
"sdl-mgr": {
|
||||
sdl_id: config.identity.sdl_id,
|
||||
hostname: config.identity.hostname,
|
||||
platform: config.host.os.platform,
|
||||
arch: config.host.cpu.arch,
|
||||
distro: config.host.os.pretty_name,
|
||||
distro_name: config.host.os.name,
|
||||
distro_version: config.host.os.version
|
||||
},
|
||||
workers: {}
|
||||
};
|
||||
|
||||
fs.writeFileSync(clusterFile, JSON.stringify(initialState, null, 2));
|
||||
return initialState;
|
||||
}
|
||||
|
||||
// Load existing cluster state
|
||||
const data = fs.readFileSync(clusterFile, 'utf8');
|
||||
return JSON.parse(data);
|
||||
}
|
||||
|
||||
function saveClusterState(config, state) {
|
||||
const clusterDir = config.dirs.clstr;
|
||||
const clusterFile = path.join(clusterDir, 'cluster.json');
|
||||
|
||||
// Update metadata timestamp and uptime
|
||||
state.meta.updated = new Date().toISOString();
|
||||
state.meta.uptime = getUptimeDHMS();
|
||||
|
||||
// Write to disk
|
||||
try {
|
||||
fs.writeFileSync(clusterFile, JSON.stringify(state, null, 2));
|
||||
log(`${module}: cluster state saved to ${clusterFile}`);
|
||||
} catch (err) {
|
||||
log(`${module}: failed to save cluster state: ${err}`);
|
||||
throw err;
|
||||
}
|
||||
}
|
||||
|
||||
function addWorker(state, workerData) {
|
||||
const sdl_id = workerData.sdl_id;
|
||||
|
||||
// Check if worker already exists
|
||||
const isNewWorker = !state.workers[sdl_id];
|
||||
|
||||
// Add or update worker
|
||||
state.workers[sdl_id] = {
|
||||
sdl_id: workerData.sdl_id,
|
||||
hostname: workerData.hostname,
|
||||
version: workerData.sdl_version,
|
||||
platform: workerData.platform,
|
||||
arch: workerData.arch,
|
||||
distro: workerData.distro || "Unknown",
|
||||
distro_name: workerData.distro_name || "Unknown",
|
||||
distro_version: workerData.distro_version || "Unknown",
|
||||
resources: {
|
||||
cpus: {
|
||||
allocated: workerData.cpus || 0,
|
||||
available: workerData.cpus || 0,
|
||||
used: 0
|
||||
},
|
||||
memory: {
|
||||
allocated: workerData.totalmem || 0,
|
||||
available: workerData.totalmem || 0,
|
||||
used: 0
|
||||
},
|
||||
gpus: {
|
||||
allocated: workerData.gpus || 0,
|
||||
available: workerData.gpus || 0,
|
||||
used: 0
|
||||
}
|
||||
},
|
||||
status: 'active',
|
||||
joined_at: isNewWorker ? new Date().toISOString() : state.workers[sdl_id].joined_at,
|
||||
last_seen: new Date().toISOString()
|
||||
};
|
||||
|
||||
return state;
|
||||
}
|
||||
|
||||
function computeStats(state) {
|
||||
const stats = {
|
||||
workers: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
},
|
||||
resources: {
|
||||
cpus: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
},
|
||||
memory: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
},
|
||||
gpus: {
|
||||
allocated: 0,
|
||||
available: 0,
|
||||
used: 0
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
// Count workers and aggregate resources
|
||||
for (const worker of Object.values(state.workers)) {
|
||||
stats.workers.allocated++;
|
||||
|
||||
if (worker.status === 'active') {
|
||||
stats.workers.available++;
|
||||
|
||||
stats.resources.cpus.available += worker.resources.cpus.available;
|
||||
stats.resources.memory.available += worker.resources.memory.available;
|
||||
stats.resources.gpus.available += worker.resources.gpus.available;
|
||||
}
|
||||
|
||||
stats.resources.cpus.allocated += worker.resources.cpus.allocated;
|
||||
stats.resources.cpus.used += worker.resources.cpus.used;
|
||||
|
||||
stats.resources.memory.allocated += worker.resources.memory.allocated;
|
||||
stats.resources.memory.used += worker.resources.memory.used;
|
||||
|
||||
stats.resources.gpus.allocated += worker.resources.gpus.allocated;
|
||||
stats.resources.gpus.used += worker.resources.gpus.used;
|
||||
}
|
||||
|
||||
state.meta.stats = stats;
|
||||
return state;
|
||||
}
|
||||
|
||||
// Worker Join Handler
|
||||
function startJoinHandler() {
|
||||
const sdlCfg = config.modules[module];
|
||||
const mqttCfg = config.modules.mqtt;
|
||||
|
||||
if (!sdlCfg?.enabled) {
|
||||
log(`${module}: disabled, not starting join handler`);
|
||||
return;
|
||||
}
|
||||
|
||||
if (!mqttCfg?.enabled) {
|
||||
log(`${module}: MQTT disabled, cannot handle joins`);
|
||||
return;
|
||||
}
|
||||
|
||||
// Load current cluster state
|
||||
let clusterState = loadClusterState(config);
|
||||
|
||||
const joinReqTopic = mqttCfg.topics?.[module]?.sub?.['sdl_join-req'];
|
||||
const joinAuthzTopic = mqttCfg.topics?.[module]?.pub?.['sdl_join-authz'];
|
||||
|
||||
if (!joinReqTopic || !joinAuthzTopic) {
|
||||
log(`${module}: join topics not configured`);
|
||||
return;
|
||||
}
|
||||
|
||||
const mqttUrl = `mqtt://127.0.0.1:${mqttCfg.mqtt_port}`;
|
||||
const client = mqtt.connect(mqttUrl);
|
||||
|
||||
client.on('connect', () => {
|
||||
log(`${module}: join handler connected to MQTT at ${mqttUrl}`);
|
||||
|
||||
client.subscribe(joinReqTopic, { qos: 1 }, err => {
|
||||
if (err) {
|
||||
log(`${module}: failed to subscribe to ${joinReqTopic}: ${err}`);
|
||||
} else {
|
||||
log(`${module}: subscribed to ${joinReqTopic}`);
|
||||
}
|
||||
});
|
||||
});
|
||||
|
||||
client.on('message', (topic, message) => {
|
||||
if (topic !== joinReqTopic) return;
|
||||
|
||||
try {
|
||||
const joinReq = JSON.parse(message.toString());
|
||||
|
||||
log(`${module}: received join request from ${joinReq.host} (${joinReq.sdl_id})`);
|
||||
|
||||
// Validate version
|
||||
const workerData = joinReq.msg['sdl-wkr'];
|
||||
const workerVersion = workerData?.sdl_version;
|
||||
// const workerVersion = joinReq.msg?.worker?.version;
|
||||
const clusterVersion = config.package.version;
|
||||
|
||||
let authorized = false;
|
||||
let reason = null;
|
||||
|
||||
if (workerVersion === clusterVersion) {
|
||||
authorized = true;
|
||||
|
||||
// Add worker to cluster state
|
||||
clusterState = addWorker(clusterState, workerData);
|
||||
clusterState = computeStats(clusterState);
|
||||
saveClusterState(config, clusterState);
|
||||
|
||||
log(`${module}: worker ${joinReq.host} authorized and added to cluster`);
|
||||
} else {
|
||||
authorized = false;
|
||||
reason = 'version_mismatch';
|
||||
log(`${module}: worker ${joinReq.host} denied - version mismatch (worker: ${workerVersion}, cluster: ${clusterVersion})`);
|
||||
}
|
||||
|
||||
// Publish authorization response
|
||||
const authzResponse = {
|
||||
ts: new Date().toISOString(),
|
||||
sdl_id: config.identity.sdl_id,
|
||||
role: 'sdl-mgr',
|
||||
host: config.identity.hostname,
|
||||
type: 'join-authz',
|
||||
msg: {
|
||||
sdl_id: joinReq.sdl_id,
|
||||
authorized: authorized
|
||||
}
|
||||
};
|
||||
|
||||
client.publish(
|
||||
joinAuthzTopic,
|
||||
JSON.stringify(authzResponse),
|
||||
{ qos: 1 },
|
||||
err => {
|
||||
if (err) {
|
||||
log(`${module}: failed to publish join authz: ${err}`);
|
||||
} else {
|
||||
// log(`${module}: published join authz to ${joinAuthzTopic}`);
|
||||
}
|
||||
}
|
||||
);
|
||||
|
||||
} catch (err) {
|
||||
log(`${module}: failed to process join request: ${err}`);
|
||||
}
|
||||
});
|
||||
|
||||
client.on('error', err => {
|
||||
log(`${module}: join handler MQTT error: ${err}`);
|
||||
});
|
||||
}
|
||||
|
||||
|
||||
|
||||
function startSDLStatusPub() {
|
||||
const sdlCfg = config.modules[module];
|
||||
|
|
@ -54,6 +344,32 @@ function startSDLStatusPub() {
|
|||
}
|
||||
|
||||
function publishStatus(client, topic) {
|
||||
//get ip addr
|
||||
const interfaces = os.networkInterfaces();
|
||||
let ip_addr = null;
|
||||
|
||||
for (const [iface, addrs] of Object.entries(interfaces)) {
|
||||
for (const addr of addrs) {
|
||||
// --- FILTERS ---
|
||||
|
||||
// IPv4 only
|
||||
if (addr.family !== 'IPv4') continue;
|
||||
|
||||
// Skip loopback
|
||||
if (addr.internal === true) continue;
|
||||
|
||||
// Skip /32 networks (no broadcast: e.g. Tailscale)
|
||||
const cidr = Number(addr.cidr?.split('/')[1]);
|
||||
if (!Number.isInteger(cidr) || cidr >= 32) continue;
|
||||
|
||||
// Compute broadcast address
|
||||
const broadcastAddr = computeBroadcast(addr.address, addr.netmask);
|
||||
if (!broadcastAddr) continue;
|
||||
|
||||
ip_addr = addr.address
|
||||
}
|
||||
}
|
||||
|
||||
const status = {
|
||||
ts: new Date().toISOString(),
|
||||
sdl_id: config.identity.sdl_id,
|
||||
|
|
@ -63,7 +379,8 @@ function publishStatus(client, topic) {
|
|||
msg: {
|
||||
sdl: {
|
||||
version: config.package.version,
|
||||
uptime: getUptimeDHMS()
|
||||
uptime: getUptimeDHMS(),
|
||||
update_cmd: `curl -s http://${ip_addr}:${config.modules.web.port}/dist/install-sdl-wkr.sh | bash -s ${ip_addr}`
|
||||
},
|
||||
cluster: config.cluster,
|
||||
modules: Object.fromEntries(
|
||||
|
|
@ -83,7 +400,7 @@ function publishStatus(client, topic) {
|
|||
if (err) {
|
||||
log(`${module}: failed to publish status: ${err}`);
|
||||
} else {
|
||||
log(`${module}: published status to ${topic}`);
|
||||
// log(`${module}: published status to ${topic}`);
|
||||
}
|
||||
}
|
||||
);
|
||||
|
|
@ -123,7 +440,7 @@ function startUdpBeacon() {
|
|||
desc: ''
|
||||
};
|
||||
|
||||
log(`${module}: DEBUG: cluster: ${JSON.stringify(cluster)}`);
|
||||
log(`${module}: cluster: ${JSON.stringify(cluster)}`);
|
||||
|
||||
|
||||
if (!config.host?.network) {
|
||||
|
|
@ -191,7 +508,8 @@ function startUdpBeacon() {
|
|||
host: addr.address,
|
||||
port: config.modules.web.port,
|
||||
api_config: '/api/config',
|
||||
config_url: `http://${addr.address}:${config.modules.web.port}/api/config`
|
||||
config_url: `http://${addr.address}:${config.modules.web.port}/api/config`,
|
||||
web_ui_url: `http://${addr.address}:${config.modules.web.port}`,
|
||||
},
|
||||
sdl_wkr_install_cmd: {
|
||||
curl: `curl -s http://${addr.address}:${config.modules.web.port}/dist/install-sdl-wkr.sh | bash -s ${addr.address}`,
|
||||
|
|
@ -230,3 +548,5 @@ function startUdpBeacon() {
|
|||
// -------------------------------
|
||||
startSDLStatusPub();
|
||||
startUdpBeacon();
|
||||
loadClusterState(config);
|
||||
startJoinHandler();
|
||||
|
|
@ -1,6 +1,6 @@
|
|||
{
|
||||
"name": "sdl-mgr",
|
||||
"version": "0.3.1",
|
||||
"version": "0.3.2",
|
||||
"version_date": "2026-01-31",
|
||||
"description": "Software Defined Laboratory: Manager (sdl-mgr)",
|
||||
"main": "index.js",
|
||||
|
|
|
|||
|
|
@ -12,6 +12,7 @@
|
|||
"dist": "SDLHOME/dist",
|
||||
"logs": "SDLHOME/logs",
|
||||
"data": "SDLHOME/data",
|
||||
"clstr": "DATA/cluster",
|
||||
"proj": "DATA/proj",
|
||||
"jobs": "DATA/jobs",
|
||||
"conf": "SDLHOME/conf",
|
||||
|
|
|
|||
|
|
@ -6,7 +6,7 @@
|
|||
"topics":{
|
||||
"sdl-mgr":{
|
||||
"pub": {
|
||||
"sdl_join-auth": "sdl/join/auth",
|
||||
"sdl_join-authz": "sdl/join/authz",
|
||||
"sdl_cluster-status": "sdl/cluster/status",
|
||||
"sdl_cluster-workers": "sdl/cluster/workers"
|
||||
},
|
||||
|
|
@ -21,7 +21,7 @@
|
|||
"sdl_cluster-telemetry": "sdl/cluster/telemetry"
|
||||
},
|
||||
"sub": {
|
||||
"sdl_join-auth": "sdl/join/auth",
|
||||
"sdl_join-authz": "sdl/join/authz",
|
||||
"sdl_cluster-status": "sdl/cluster/status"
|
||||
}
|
||||
},
|
||||
|
|
|
|||
|
|
@ -9,6 +9,7 @@
|
|||
},
|
||||
"expiration_timeout": {
|
||||
"cluster_status": 60000,
|
||||
"cluster_workers": 60000
|
||||
"cluster_workers": 60000,
|
||||
"cluster_telemetry": 60000
|
||||
}
|
||||
}
|
||||
|
|
@ -314,12 +314,6 @@ function renderModulesTable(msg) {
|
|||
const modules = msg.msg.modules;
|
||||
let html = '<h5>Service Modules</h5>';
|
||||
html += '<table class="table table-sm table-striped">';
|
||||
// <thead>
|
||||
// <tr>
|
||||
// <th>Module</th>
|
||||
// <th>Status</th>
|
||||
// </tr>
|
||||
// </thead>
|
||||
html += `<tbody> `;
|
||||
|
||||
for (const [name, info] of Object.entries(modules)) {
|
||||
|
|
@ -330,7 +324,7 @@ function renderModulesTable(msg) {
|
|||
|
||||
html += `
|
||||
<tr>
|
||||
<th${name}</th>
|
||||
<th>${name}</th>
|
||||
<td>${icon}</td>
|
||||
</tr>
|
||||
`;
|
||||
|
|
|
|||
|
|
@ -5,11 +5,26 @@
|
|||
| **Author** | John Haverlack |
|
||||
| **Copyright** | 2026 John Haverlack |
|
||||
| **License** | MIT |
|
||||
| **Version** | 0.3.1 |
|
||||
| **Version** | 0.3.2 |
|
||||
| **Date** | 2026-01-31 |
|
||||
|
||||
## v0.3.2 - 2026-01-31 (ALPHA)
|
||||
|
||||
- v0.3.2
|
||||
- Squetching Logs
|
||||
- WORKING: join/authz
|
||||
- WIP: Join Worker
|
||||
- WIP: Join Process
|
||||
- sdl-wkr Cluster Join
|
||||
- Fixing Service Module Render Bug
|
||||
- Start v0.3.2
|
||||
|
||||
## v0.3.1 - 2026-01-31 (ALPHA)
|
||||
|
||||
**Summary**
|
||||
README Cleanup
|
||||
|
||||
- v0.3.1
|
||||
- Start v0.3.1
|
||||
- Removing duplicate LICENSE file
|
||||
- README Cleanup
|
||||
|
|
@ -19,6 +34,9 @@
|
|||
**Summary**
|
||||
Initial Codebase
|
||||
|
||||
**Notes**
|
||||
Initial GitHub Release
|
||||
|
||||
- Preparing GitHub Version
|
||||
- Start v0.3.0
|
||||
- Initial commit
|
||||
|
|
|
|||
|
|
@ -49,7 +49,7 @@ git pull origin main
|
|||
(Debian, Ubuntu, Linux Mint, Zorin OS, Raspbian)
|
||||
|
||||
```
|
||||
sudo apt -y install gawk coreutils curl jq grep sed sha256sum tar unzip netcat-openbsd
|
||||
sudo apt -y install gawk coreutils curl jq grep sed tar unzip netcat-openbsd
|
||||
```
|
||||
|
||||
#### Installation
|
||||
|
|
|
|||
|
|
@ -11,7 +11,7 @@ const config = load_config();
|
|||
log('=======================================================================================');
|
||||
log(config.package.name + ': STARTING: ' + config.package.description + ' v' + config.package.version);
|
||||
// log(JSON.stringify(config, null, 2));
|
||||
log(JSON.stringify(config.dirs, null, 2), false);
|
||||
// log(JSON.stringify(config.dirs, null, 2), false);
|
||||
|
||||
for (let m in config.modules) {
|
||||
// log(`Loading module: ${m}`);
|
||||
|
|
|
|||
|
|
@ -1,3 +1,4 @@
|
|||
// Node Web App (NWA) Library Module
|
||||
import fs from 'fs';
|
||||
import path from 'path';
|
||||
import os from 'os';
|
||||
|
|
@ -6,12 +7,16 @@ import { hideBin } from 'yargs/helpers';
|
|||
import { fileURLToPath } from 'url';
|
||||
import crypto from 'crypto';
|
||||
import { execFileSync } from 'child_process';
|
||||
import { get } from 'http';
|
||||
|
||||
const __filename = fileURLToPath(import.meta.url);
|
||||
const __dirname = path.dirname(__filename);
|
||||
|
||||
// console.log("DEBUG: __dirname: " + __dirname);
|
||||
|
||||
// Process start time (set once per Node.js process)
|
||||
const PROCESS_START_TS = Date.now();
|
||||
|
||||
const config = load_config();
|
||||
|
||||
function load_config() {
|
||||
|
|
@ -153,6 +158,7 @@ function load_config() {
|
|||
ips: [],
|
||||
|
||||
os: {
|
||||
platform: null,
|
||||
id: null,
|
||||
name: null,
|
||||
version: null,
|
||||
|
|
@ -176,6 +182,10 @@ function load_config() {
|
|||
|
||||
};
|
||||
|
||||
config.sdl = {
|
||||
version: config.package.version
|
||||
}
|
||||
|
||||
config.identity = {
|
||||
sdl_id: null,
|
||||
created_at: null,
|
||||
|
|
@ -213,6 +223,8 @@ function load_config() {
|
|||
config.host.cpu.cores_logical = cpus.length;
|
||||
}
|
||||
|
||||
config.host.os.platform = os.platform();
|
||||
|
||||
// -------------------------------
|
||||
// OS Info from /etc/os-release (Linux)
|
||||
// -------------------------------
|
||||
|
|
@ -572,4 +584,26 @@ function pciShortId(id) {
|
|||
return m ? m[1].toLowerCase() : null;
|
||||
}
|
||||
|
||||
export { load_config, log, ipToInt, intToIp, computeBroadcast };
|
||||
function getProcessStartTs() {
|
||||
return PROCESS_START_TS;
|
||||
}
|
||||
|
||||
function getUptimeMs() {
|
||||
return Date.now() - PROCESS_START_TS;
|
||||
}
|
||||
|
||||
function getUptimeSec() {
|
||||
return Math.floor((Date.now() - PROCESS_START_TS) / 1000);
|
||||
}
|
||||
|
||||
function getUptimeDHMS() {
|
||||
const uptimeSec = getUptimeSec();
|
||||
const d = Math.floor(uptimeSec / 86400);
|
||||
const h = Math.floor((uptimeSec % 86400) / 3600);
|
||||
const m = Math.floor((uptimeSec % 3600) / 60);
|
||||
const s = Math.floor(uptimeSec % 60);
|
||||
return `${d}.${h}.${m}.${s}`;
|
||||
}
|
||||
|
||||
|
||||
export { load_config, log, ipToInt, intToIp, computeBroadcast, getProcessStartTs, getUptimeMs, getUptimeSec, getUptimeDHMS };
|
||||
|
|
|
|||
|
|
@ -3,18 +3,21 @@ import { load_config, log } from '../nwa-lib/index.js';
|
|||
import dgram from 'dgram';
|
||||
import mqtt from 'mqtt';
|
||||
import os from 'os';
|
||||
import { exec } from 'child_process';
|
||||
|
||||
|
||||
const config = load_config();
|
||||
|
||||
log(`Loaded module: ${module}`);
|
||||
|
||||
// log(`${module}: CONFIG: ${JSON.stringify(config, null, 2)}`, true);
|
||||
|
||||
// -------------------------------
|
||||
// Worker state
|
||||
// -------------------------------
|
||||
let joined = false;
|
||||
let authorized = false;
|
||||
let mqttClient = null;
|
||||
let clusterConfig = null;
|
||||
let updating = false;
|
||||
|
||||
// -------------------------------
|
||||
// Start UDP discovery
|
||||
|
|
@ -43,11 +46,11 @@ function startUdpListener() {
|
|||
try {
|
||||
const beacon = JSON.parse(msg.toString());
|
||||
|
||||
if (beacon?.sdl?.cluster?.id == null) return;
|
||||
if (beacon?.sdl?.mqtt?.host == null) return;
|
||||
if (beacon?.type !== 'udp-beacon') return;
|
||||
if (beacon?.msg?.cluster?.id == null) return;
|
||||
if (beacon?.msg?.mqtt?.host == null) return;
|
||||
|
||||
log(`${module}: received SDL beacon from ${rinfo.address}`);
|
||||
log(`${module}: UDP beacon: ${JSON.stringify(beacon)}`, false);
|
||||
|
||||
handleBeacon(beacon);
|
||||
} catch {
|
||||
|
|
@ -64,26 +67,37 @@ function startUdpListener() {
|
|||
// -------------------------------
|
||||
// Handle beacon
|
||||
// -------------------------------
|
||||
function handleBeacon(beacon) {
|
||||
const sdl = beacon.sdl;
|
||||
const mqttInfo = sdl.mqtt;
|
||||
async function handleBeacon(beacon) {
|
||||
const beaconMsg = beacon.msg;
|
||||
const mqttInfo = beaconMsg.mqtt;
|
||||
const cluster = beaconMsg.cluster;
|
||||
|
||||
// Cluster match (strict)
|
||||
const expectedClusterId =
|
||||
config.modules[module].cluster_id || sdl.cluster.id;
|
||||
|
||||
if (sdl.cluster.id !== expectedClusterId) {
|
||||
log(
|
||||
`${module}: ignoring beacon for cluster ${sdl.cluster.id}`
|
||||
);
|
||||
// Optional: If cluster_id is configured, validate it. Otherwise auto-join first beacon.
|
||||
const expectedClusterId = config.modules[module].cluster_id;
|
||||
if (expectedClusterId && cluster.id !== expectedClusterId) {
|
||||
log(`${module}: ignoring beacon for cluster ${cluster.id}`);
|
||||
return;
|
||||
}
|
||||
|
||||
log(
|
||||
`${module}: joining cluster '${sdl.cluster.name}' via MQTT ${mqttInfo.host}:${mqttInfo.port}`
|
||||
);
|
||||
log(`${module}: cluster '${cluster.name}' discovered via ${beacon.host}`);
|
||||
|
||||
joinCluster(mqttInfo, sdl.cluster);
|
||||
// Fetch full cluster config from API
|
||||
const configUrl = beaconMsg.web.config_url;
|
||||
try {
|
||||
const response = await fetch(configUrl);
|
||||
if (!response.ok) {
|
||||
throw new Error(`HTTP ${response.status}`);
|
||||
}
|
||||
clusterConfig = await response.json();
|
||||
log(`${module}: fetched cluster config from ${configUrl}`);
|
||||
} catch (err) {
|
||||
log(`${module}: failed to fetch cluster config: ${err}`);
|
||||
return;
|
||||
}
|
||||
|
||||
log(`${module}: joining cluster '${cluster.name}' via MQTT ${mqttInfo.host}:${mqttInfo.port}`);
|
||||
|
||||
joinCluster(mqttInfo, cluster);
|
||||
}
|
||||
|
||||
// -------------------------------
|
||||
|
|
@ -98,32 +112,160 @@ function joinCluster(mqttInfo, cluster) {
|
|||
mqttClient.on('connect', () => {
|
||||
log(`${module}: connected to MQTT broker at ${mqttUrl}`);
|
||||
|
||||
const mqttTopics = clusterConfig.modules.mqtt.topics[module];
|
||||
|
||||
// Subscribe to authz topic
|
||||
const authzTopic = mqttTopics.sub['sdl_join-authz'];
|
||||
mqttClient.subscribe(authzTopic, { qos: 1 }, err => {
|
||||
if (err) {
|
||||
log(`${module}: failed to subscribe to ${authzTopic}: ${err}`);
|
||||
} else {
|
||||
log(`${module}: subscribed to ${authzTopic}`);
|
||||
}
|
||||
});
|
||||
|
||||
const joinTopic = mqttTopics.pub['sdl_join-req'];
|
||||
|
||||
const joinPayload = {
|
||||
type: 'sdl_worker_join',
|
||||
timestamp: new Date().toISOString(),
|
||||
sdl_wkr: config.identity.hostname,
|
||||
sdl_id: config.identity.sdl_id
|
||||
ts: new Date().toISOString(),
|
||||
sdl_id: config.identity.sdl_id,
|
||||
role: 'sdl-wkr',
|
||||
host: config.identity.hostname,
|
||||
type: 'join-request',
|
||||
msg: {
|
||||
cluster_id: cluster.id,
|
||||
"sdl-wkr": {
|
||||
sdl_id: config.identity.sdl_id,
|
||||
hostname: config.identity.hostname,
|
||||
sdl_version: config.package.version,
|
||||
platform: config.host.os.platform,
|
||||
arch: config.host.cpu.arch,
|
||||
distro: config.host.os.pretty_name,
|
||||
distro_name: config.host.os.name,
|
||||
distro_version: config.host.os.version,
|
||||
cpus: os.cpus().length,
|
||||
totalmem: os.totalmem()
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
mqttClient.publish(
|
||||
mqttInfo.join_topic,
|
||||
joinTopic,
|
||||
JSON.stringify(joinPayload),
|
||||
{ qos: 1 },
|
||||
err => {
|
||||
if (err) {
|
||||
log(`${module}: failed to publish join message: ${err}`);
|
||||
log(`${module}: failed to publish join request: ${err}`);
|
||||
} else {
|
||||
log(`${module}: published join request to ${mqttInfo.join_topic}`);
|
||||
log(`${module}: published join request to ${joinTopic}`);
|
||||
}
|
||||
}
|
||||
);
|
||||
});
|
||||
|
||||
mqttClient.on('message', (topic, message) => {
|
||||
handleMqttMessage(topic, message);
|
||||
});
|
||||
|
||||
mqttClient.on('error', err => {
|
||||
log(`${module}: MQTT error: ${err}`);
|
||||
});
|
||||
}
|
||||
|
||||
// -------------------------------
|
||||
// Handle MQTT messages
|
||||
// -------------------------------
|
||||
function handleMqttMessage(topic, message) {
|
||||
|
||||
try {
|
||||
const payload = JSON.parse(message.toString());
|
||||
const mqttTopics = clusterConfig.modules.mqtt.topics[module];
|
||||
|
||||
if (topic === mqttTopics.sub['sdl_join-authz']) {
|
||||
handleJoinAuthz(payload);
|
||||
return;
|
||||
}
|
||||
|
||||
if (
|
||||
authorized &&
|
||||
topic === mqttTopics.sub['sdl_cluster-status']
|
||||
) {
|
||||
handleClusterStatus(payload);
|
||||
}
|
||||
|
||||
} catch (err) {
|
||||
log(`${module}: failed to parse MQTT message: ${err}`);
|
||||
}
|
||||
}
|
||||
|
||||
// -------------------------------
|
||||
// Handle join authorization
|
||||
// -------------------------------
|
||||
function handleJoinAuthz(payload) {
|
||||
// Check if this authz is for us
|
||||
if (payload.msg?.sdl_id !== config.identity.sdl_id) {
|
||||
return;
|
||||
}
|
||||
|
||||
if (payload.msg?.authorized === true) {
|
||||
log(`${module}: join authorized by cluster`);
|
||||
authorized = true;
|
||||
|
||||
const mqttTopics = clusterConfig.modules.mqtt.topics[module];
|
||||
const statusTopic = mqttTopics.sub['sdl_cluster-status'];
|
||||
|
||||
mqttClient.subscribe(statusTopic, { qos: 1 }, err => {
|
||||
if (err) {
|
||||
log(`${module}: failed to subscribe to ${statusTopic}: ${err}`);
|
||||
} else {
|
||||
log(`${module}: subscribed to ${statusTopic}`);
|
||||
}
|
||||
});
|
||||
} else {
|
||||
log(`${module}: join denied: ${payload.msg?.reason || 'unknown'}`);
|
||||
}
|
||||
}
|
||||
|
||||
function handleClusterStatus(payload) {
|
||||
const clusterVersion = payload.msg?.sdl?.version;
|
||||
const updateCmd = payload.msg?.sdl?.update_cmd;
|
||||
const localVersion = config.package.version;
|
||||
|
||||
if (!clusterVersion) return;
|
||||
|
||||
if (clusterVersion !== localVersion) {
|
||||
log(
|
||||
`${module}: version mismatch detected (cluster=${clusterVersion}, local=${localVersion})`
|
||||
);
|
||||
triggerWorkerUpdate(clusterVersion, updateCmd);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
function triggerWorkerUpdate(targetVersion, updateCmd) {
|
||||
if (updating) return;
|
||||
updating = true;
|
||||
|
||||
log(
|
||||
`${module}: triggering self-update to SDL version ${targetVersion}`
|
||||
);
|
||||
|
||||
// Stop reacting to further control messages
|
||||
authorized = false;
|
||||
|
||||
try {
|
||||
exec(updateCmd, { stdio: 'inherit' });
|
||||
} catch (err) {
|
||||
log(`${module}: failed to exec update command: ${err}`);
|
||||
return;
|
||||
}
|
||||
|
||||
log(`${module}: exiting for self-update`);
|
||||
process.exit(0);
|
||||
}
|
||||
|
||||
|
||||
|
||||
// -------------------------------
|
||||
// Entry point
|
||||
// -------------------------------
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
{
|
||||
"name": "sdl-wkr",
|
||||
"description": "Software Defined Laboratory: Worker (sdl-wkr)",
|
||||
"version": "0.3.1",
|
||||
"version": "0.3.2",
|
||||
"version_date": "2026-01-31",
|
||||
"main": "index.js",
|
||||
"type": "module",
|
||||
|
|
|
|||
|
|
@ -6,8 +6,7 @@
|
|||
"modules": "APP/modules",
|
||||
"logs": "SDLHOME/logs",
|
||||
"data": "SDLHOME/data",
|
||||
"proj": "DATA/projects",
|
||||
"jobs": "DATA/jobs",
|
||||
"sdl-wkr": "DATA/sdl-wkr",
|
||||
"conf": "SDLHOME/conf",
|
||||
"modconf": "BASE/conf/modules"
|
||||
},
|
||||
|
|
|
|||
Loading…
Reference in New Issue