Hua/nanobot
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: unify instance path resolution and preserve workspace override

Browse Source
This commit is contained in:
Re-bin
2026-03-08 02:58:25 +00:00
parent bdac08161b
commit 20dfaa5d34
16 changed files with 282 additions and 210 deletions
Download Patch File Download Diff File Expand all files Collapse all files

192
README.md
View File

@@ -905,7 +905,7 @@ MCP tools are automatically discovered and registered on startup. The LLM can us
## Multiple Instances
Run multiple nanobot instances simultaneously with complete isolation. Each instance has its own configuration, workspace, cron jobs, logs, and media storage.
Run multiple nanobot instances simultaneously with separate configs and runtime data. Use `--config` as the main entrypoint, and optionally use `--workspace` to override the workspace for a specific run.
### Quick Start
@@ -920,35 +920,31 @@ nanobot gateway --config ~/.nanobot-discord/config.json
nanobot gateway --config ~/.nanobot-feishu/config.json --port 18792
```
### Complete Isolation
### Path Resolution
When using `--config` parameter, nanobot automatically derives the data directory from the config file path, ensuring complete isolation:
When using `--config`, nanobot derives its runtime data directory from the config file location. The workspace still comes from `agents.defaults.workspace` unless you override it with `--workspace`.
| Component | Isolation | Example |
|-----------|-----------|---------|
| **Config** | Separate config files | `~/.nanobot-A/config.json`, `~/.nanobot-B/config.json` |
| **Workspace** | Independent memory, sessions, skills | `~/.nanobot-A/workspace/`, `~/.nanobot-B/workspace/` |
| **Cron Jobs** | Separate job storage | `~/.nanobot-A/cron/`, `~/.nanobot-B/cron/` |
| **Logs** | Independent log files | `~/.nanobot-A/logs/`, `~/.nanobot-B/logs/` |
| **Media** | Separate media storage | `~/.nanobot-A/media/`, `~/.nanobot-B/media/` |
| Component | Resolved From | Example |
|-----------|---------------|---------|
| **Config** | `--config` path | `~/.nanobot-A/config.json` |
| **Workspace** | `--workspace` or config | `~/.nanobot-A/workspace/` |
| **Cron Jobs** | config directory | `~/.nanobot-A/cron/` |
| **Media / runtime state** | config directory | `~/.nanobot-A/media/` |
### Setup Example
### How It Works
**1. Create directory structure for each instance:**
- `--config` selects which config file to load
- By default, the workspace comes from `agents.defaults.workspace` in that config
- If you pass `--workspace`, it overrides the workspace from the config file
```bash
# Instance A
mkdir -p ~/.nanobot-telegram/{workspace,cron,logs,media}
cp ~/.nanobot/config.json ~/.nanobot-telegram/config.json
### Minimal Setup
# Instance B
mkdir -p ~/.nanobot-discord/{workspace,cron,logs,media}
cp ~/.nanobot/config.json ~/.nanobot-discord/config.json
```
1. Copy your base config into a new instance directory.
2. Set a different `agents.defaults.workspace` for that instance.
3. Start the instance with `--config`.
**2. Configure each instance:**
Example config:
Edit `~/.nanobot-telegram/config.json`:
```json
{
"agents": {
@@ -969,160 +965,32 @@ Edit `~/.nanobot-telegram/config.json`:
}
```
Edit `~/.nanobot-discord/config.json`:
```json
{
"agents": {
"defaults": {
"workspace": "~/.nanobot-discord/workspace",
"model": "anthropic/claude-opus-4"
}
},
"channels": {
"discord": {
"enabled": true,
"token": "YOUR_DISCORD_BOT_TOKEN"
}
},
"gateway": {
"port": 18791
}
}
```
**3. Start instances:**
Start separate instances:
```bash
# Terminal 1
nanobot gateway --config ~/.nanobot-telegram/config.json
# Terminal 2
nanobot gateway --config ~/.nanobot-discord/config.json
```
### Use Cases
- **Multiple Chat Platforms**: Run separate bots for Telegram, Discord, Feishu, etc.
- **Different Models**: Test different LLM models (Claude, GPT, DeepSeek) simultaneously
- **Role Separation**: Dedicated instances for different purposes (personal assistant, work bot, research agent)
- **Multi-Tenant**: Serve multiple users/teams with isolated configurations
### Management Scripts
For production deployments, create management scripts for each instance:
Override workspace for one-off runs when needed:
```bash
#!/bin/bash
# manage-telegram.sh
INSTANCE_NAME="telegram"
CONFIG_FILE="$HOME/.nanobot-telegram/config.json"
LOG_FILE="$HOME/.nanobot-telegram/logs/stderr.log"
case "$1" in
start)
nohup nanobot gateway --config "$CONFIG_FILE" >> "$LOG_FILE" 2>&1 &
echo "Started $INSTANCE_NAME instance (PID: $!)"
;;
stop)
pkill -f "nanobot gateway.*$CONFIG_FILE"
echo "Stopped $INSTANCE_NAME instance"
;;
restart)
$0 stop
sleep 2
$0 start
;;
status)
pgrep -f "nanobot gateway.*$CONFIG_FILE" > /dev/null
if [ $? -eq 0 ]; then
echo "$INSTANCE_NAME instance is running"
else
echo "$INSTANCE_NAME instance is not running"
fi
;;
*)
echo "Usage: $0 {start|stop|restart|status}"
exit 1
;;
esac
nanobot gateway --config ~/.nanobot-telegram/config.json --workspace /tmp/nanobot-telegram-test
```
### systemd Service (Linux)
### Common Use Cases
For automatic startup and crash recovery, create a systemd service for each instance:
```ini
# ~/.config/systemd/user/nanobot-telegram.service
[Unit]
Description=Nanobot Telegram Instance
After=network.target
[Service]
Type=simple
ExecStart=%h/.local/bin/nanobot gateway --config %h/.nanobot-telegram/config.json
Restart=always
RestartSec=10
[Install]
WantedBy=default.target
```
Enable and start:
```bash
systemctl --user daemon-reload
systemctl --user enable --now nanobot-telegram
systemctl --user enable --now nanobot-discord
```
### launchd Service (macOS)
Create a plist file for each instance:
```xml
<!-- ~/Library/LaunchAgents/com.nanobot.telegram.plist -->
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>Label</key>
<string>com.nanobot.telegram</string>
<key>ProgramArguments</key>
<array>
<string>/path/to/nanobot</string>
<string>gateway</string>
<string>--config</string>
<string>/Users/yourname/.nanobot-telegram/config.json</string>
</array>
<key>RunAtLoad</key>
<true/>
<key>KeepAlive</key>
<true/>
<key>StandardOutPath</key>
<string>/Users/yourname/.nanobot-telegram/logs/stdout.log</string>
<key>StandardErrorPath</key>
<string>/Users/yourname/.nanobot-telegram/logs/stderr.log</string>
</dict>
</plist>
```
Load the service:
```bash
launchctl load ~/Library/LaunchAgents/com.nanobot.telegram.plist
launchctl load ~/Library/LaunchAgents/com.nanobot.discord.plist
```
- Run separate bots for Telegram, Discord, Feishu, and other platforms
- Keep testing and production instances isolated
- Use different models or providers for different teams
- Serve multiple tenants with separate configs and runtime data
### Notes
- Each instance must use a different port (default: 18790)
- Instances are completely independent — no shared state or cross-talk
- You can run different LLM models, providers, and channel configurations per instance
- Memory, sessions, and cron jobs are fully isolated between instances
- Each instance must use a different port if they run at the same time
- Use a different workspace per instance if you want isolated memory, sessions, and skills
- `--workspace` overrides the workspace defined in the config file
- Cron jobs and runtime media/state are derived from the config directory
## CLI Reference