OmniHSS Metrics and Monitoring Guide

← Back to Operations Guide

---

Table of Contents

• Monitoring Overview
• Control Panel Monitoring
• Database Monitoring
• Log Monitoring
• External Monitoring Integration
• Key Performance Indicators
• Alerting Strategies

---

Monitoring Overview

OmniHSS provides several mechanisms for monitoring system health, performance, and subscriber activity. Operations staff should utilize a combination of these tools for comprehensive visibility.

Monitoring Layers

---

Control Panel Monitoring

The Control Panel provides the primary real-time monitoring interface.

Overview Page Monitoring

URL: https://[hostname]:7443/overview

Key Metrics Available

Monitored Subscriber States

State	Indicator	What It Means
Idle	No location info	Subscriber powered off or out of coverage
Attached	MME present	Subscriber registered to network
PDN Active	PDN session count > 0	Active data connection
IMS Registered	S-CSCF assigned	Voice services ready
In Call	Active call count > 0	VoLTE call in progress

Extracting Metrics from Overview

While the Control Panel doesn't export metrics directly, you can:

1. Count visible rows for total subscribers
2. Scan for green checkmarks to count enabled subscribers
3. Review expanded details for state information
4. Note last seen timestamps for responsiveness

Diameter Page Monitoring

URL: https://[hostname]:7443/diameter

Key Metrics

Critical Peer Monitoring

Identify critical peers and monitor their status:

Peer Type	Criticality	Impact if Down
MME	High	No new LTE attachments
P-GW	High	No data sessions
S-CSCF	High	No IMS registrations
P-CSCF	High	No VoLTE calls
I-CSCF	Medium	IMS routing issues
AS	Low-Medium	Specific service unavailable

Application Page Monitoring

URL: https://[hostname]:7443/application

Key Metrics

Metric	Description	Normal Range	Action Threshold
Process Count	Active Erlang processes	Varies by load	> 90% of limit
Memory Usage	Total memory consumed	< 80%	> 90%
Uptime	Time since last restart	N/A	Track for stability

---

Database Monitoring

Direct Database Queries

Connect to SQL Database to extract detailed metrics:

Subscriber Counts

Query the database to retrieve:

• Total count of all subscribers
• Count of enabled subscribers
• Count of IMS-enabled subscribers

Session Statistics

Query the database to retrieve:

• Count of active PDN sessions
• Count of active VoLTE calls
• Breakdown of PDN sessions by APN profile

Location Statistics

Query the database to retrieve:

• Subscriber count grouped by visited network (MCC-MNC combination)
• Count of subscribers currently roaming (not on home PLMN 001-001)
• Distribution of subscribers across different visited networks

Recent Activity

Query the database to retrieve:

• Count of subscribers seen in the last hour
• Distribution of subscribers by serving MME
• Timestamp analysis of last subscriber activity

Database Health Monitoring

Monitor database health by querying:

• Total database size and growth trends
• Individual table sizes and row counts
• Current database connection count
• Query performance and resource usage

---

Log Monitoring

Log Output

OmniHSS outputs logs to stdout/stderr, which should be captured by your process manager.

Log Levels

Key Log Patterns to Monitor

Diameter Peer Events:

[info] Diameter peer connected: mme01.epc.example.com
[warn] Diameter peer disconnected: pgw01.epc.example.com
[error] Diameter peer connection failed: timeout

Database Events:

[info] Database connection established
[error] Database connection lost: timeout
[error] Database query failed: deadlock detected

Authentication Events:

[info] Authentication successful: IMSI 001001123456789
[warn] Authentication failed: IMSI 001001123456789, invalid vector
[error] Roaming denied: IMSI 001001123456789, MCC 310 MNC 410

Log Aggregation

For production deployments, implement log aggregation:

---

External Monitoring Integration

Health Check Endpoint

API Health Check: GET /api/status

curl -k https://hss.example.com:8443/api/status

Expected Response:

{"status": "ok"}

HTTP Status: 200 OK

Monitoring Tool Integration

Nagios/Icinga Example

#!/bin/bash
# check_omnihss.sh

API_URL="https://hss.example.com:8443/api/status"

response=$(curl -k -s -o /dev/null -w "%{http_code}" "$API_URL" --max-time 5)

if [ "$response" = "200" ]; then
    echo "OK - OmniHSS API responding"
    exit 0
else
    echo "CRITICAL - OmniHSS API not responding (HTTP $response)"
    exit 2
fi

Prometheus Integration

Custom exporters can be created to export OmniHSS metrics to Prometheus by querying the API and database.

SNMP Integration

For SNMP-based monitoring, custom SNMP extension scripts can query the database or API for metrics and return values via SNMP OIDs.

---

Key Performance Indicators

Operational KPIs

Recommended KPI Thresholds

KPI	Target	Warning	Critical
System Uptime	99.99%	< 99.95%	< 99.9%
Diameter Peer Uptime	99.9%	< 99.5%	< 99%
Authentication Success Rate	> 99%	< 99%	< 95%
Diameter Response Time	< 100ms	> 200ms	> 500ms
Database Query Time	< 50ms	> 100ms	> 500ms
Error Rate	< 0.1%	> 0.5%	> 1%

Capacity KPIs

Metric	Monitor	Plan Action At
Total Subscribers	Current count	80% of expected capacity
Concurrent PDN Sessions	Active sessions	70% of expected maximum
Database Size	MB used	80% of allocated storage
Database Connections	Active connections	80% of pool size

---

Alerting Strategies

Alert Priorities

Alert Definitions

Critical Alerts (P1)

System Unavailable:

• API health check fails
• Control Panel inaccessible
• Database connection fails
• Action: Immediate investigation and escalation

All Diameter Peers Disconnected:

• Zero connected peers
• Action: Check network, restart if necessary

Database Down:

• Cannot connect to SQL Database
• Action: Investigate database server, restart if necessary

High Priority Alerts (P2)

Critical Diameter Peer Down:

• Primary MME disconnected
• Primary P-GW disconnected
• Primary S-CSCF disconnected
• Action: Investigate peer connectivity within 15 minutes

High Memory Usage:

• Memory > 95%
• Action: Investigate memory leak, plan restart

High Authentication Failure Rate:

• 10% of auth requests fail

• Action: Check subscriber provisioning, investigate cause

Medium Priority Alerts (P3)

Non-Critical Peer Down:

• Secondary peer disconnected
• Application Server disconnected
• Action: Investigate within 1 hour

Elevated Memory Usage:

• Memory > 85%
• Action: Monitor trend, plan capacity upgrade

Elevated Error Rate:

• Error rate > 1%
• Action: Review logs, identify root cause

Low Priority Alerts (P4)

Capacity Warning:

• Subscribers > 80% of capacity
• Database > 80% of allocated storage
• Action: Plan capacity expansion

Performance Degradation:

• Response times elevated but acceptable
• Action: Monitor and optimize queries

Alert Notification Channels

---

Monitoring Checklist

Daily Checks

• Review Control Panel Overview - subscriber counts normal
• Review Diameter page - all critical peers connected
• Review Application page - memory and processes within limits
• Check for error logs - no critical errors in last 24 hours
• Verify backup completed successfully

Weekly Checks

• Review capacity trends - subscriber growth
• Review performance trends - response times
• Review database size - growth rate acceptable
• Review error rates - identify patterns
• Test alert notifications - ensure working

Monthly Checks

• Capacity planning review - project 6 months ahead
• Performance optimization review - identify slow queries
• Security review - certificate expiration, vulnerabilities
• Documentation review - update runbooks
• Disaster recovery test - verify backups restore correctly

---

← Back to Operations Guide | Next: Multi-Features →