Skip to main content
The longmem status command displays the current state of the memory daemon.

​
Syntax

longmem status
This command takes no options or arguments.

​
Output

​
When Running

$ longmem status
Daemon: running
Version: 0.3.0
Database: 2.45 MB

​
When Stopped

$ longmem status
Daemon: stopped
Version: 0.3.0

​
Information Displayed

​
Daemon Status

Indicates whether the daemon is responding to health checks:
  • running - Daemon is active and responding
  • stopped - Daemon is not running or not responding

​
Version

Shows the installed version of LongMem.

​
Database Size

When the daemon is running, shows the size of the SQLite database file (~/.longmem/longmem.db) in megabytes. Typical sizes:
  • New installation: <1 MB
  • Active use (1 week): 2-10 MB
  • Long-term use (months): 10-100 MB

​
Detailed Status

For more detailed daemon information, you can query the status endpoint directly: 
curl http://127.0.0.1:5939/status
This returns a JSON response with additional fields: 
{
  "status": "ok",
  "pid": 12345,
  "port": 5939,
  "uptime": 3600,
  "pending": 0,
  "circuit_open": false,
  "idle_ms": 45000,
  "service_managed": false,
  "version": "0.3.0"
}

​
Status Fields

FieldDescription
pidProcess ID of the daemon
portHTTP port the daemon is listening on
uptimeSeconds since daemon started
pendingNumber of compression jobs in queue
circuit_openWhether circuit breaker is open (paused due to errors)
idle_msMilliseconds since last activity
service_managedWhether running under systemd/launchd
versionLongMem version

​
Using the Daemon Status Command

You can also run the daemon’s built-in status command for even more detail: 
longmemd --status
Or: 
bun ~/.longmem/daemon.js --status
This produces output like: 
longmem daemon: running
  PID:        12345
  Port:       5939
  Uptime:     3600s
  Pending:    0 compression jobs
  Circuit:    closed
  Idle:       45s
  Service:    manual

​
Understanding the Fields

PID: Process identifier - useful for troubleshooting or manual kill Port: HTTP API port - must match configuration Uptime: How long the daemon has been running - resets on restart Pending: Compression jobs waiting to be processed:
  • 0 - All caught up
  • 1-10 - Normal during active use
  • >50 - May indicate compression is struggling or circuit breaker is open
Circuit: Circuit breaker status:
  • closed - Normal operation
  • OPEN (paused) - Too many compression failures, paused for cooldown
Idle: Time since last activity:
  • Low (0-60s) - Active use
  • High (>300s) - Idle, may trigger compression
Service: Lifecycle management:
  • manual - Started with longmem start
  • systemd/launchd - Managed by system service manager

​
Exit Codes

  • 0 - Daemon is running
  • 1 - Daemon is stopped
This allows using longmem status in scripts: 
if longmem status > /dev/null 2>&1; then
  echo "Daemon is running"
else
  echo "Daemon is stopped"
  longmem start
fi

​
Health Check Mechanism

The status command checks daemon health by sending a request to: 
GET http://127.0.0.1:5939/health
The health endpoint returns: 
{
  "status": "ok",
  "uptime": 3600
}
If this request succeeds, the daemon is considered running.

​
Troubleshooting

​
Status shows “stopped” but process is running

# Check for running processes
ps aux | grep longmemd

# Check if PID file exists
cat ~/.longmem/longmem.pid

# Check if port is listening
lsof -ti:5939

# Try daemon status directly
longmemd --status
This can happen if:
  • Daemon is running but not responding (hung)
  • Different port is configured
  • Firewall blocking localhost

​
Stale PID file

If the daemon crashes, it may leave a stale PID file: 
longmem daemon: stopped
  Stale PID file: ~/.longmem/longmem.pid (pid 12345, process dead)
Fix by removing the stale file: 
rm ~/.longmem/longmem.pid

Build docs developers (and LLMs) love

Get started for freeTalk to us