Skip to main content

Web UI Operations Guide

Table of Contents​

  1. Overview
  2. Accessing the Control Panel
  3. P-CSCF Management
  4. I-CSCF Management
  5. S-CSCF Management
  6. Diameter Peer Management
  7. Hash Table Operations
  8. Logs Viewing
  9. Monitoring and Metrics

Overview​

The OmniCall CSCF Web UI provides a comprehensive control panel for real-time monitoring and management of all CSCF components (P-CSCF, I-CSCF, S-CSCF). The interface is built on Phoenix LiveView and provides:

  • Real-time visibility into registrations, active calls, and system state
  • Hash table management for performance-critical in-memory data structures
  • Diameter peer monitoring and control
  • Prometheus metrics for system monitoring
  • Live log viewing for troubleshooting

Architecture​

The control panel communicates with CSCF backend instances to:

  • Query user registrations and location data
  • Inspect active dialogs (calls)
  • Manage Diameter peers
  • View and manipulate hash tables
  • Access Initial Filter Criteria (IFC) configuration

Accessing the Control Panel​

Default Access​

The control panel is accessible via HTTP on the CSCF server:

http://<cscf-server>:4000/

Default Port: 4000 (configurable in ControlPanel.Supervisor)

Configuration​

The control panel requires CSCF host configuration in config/config.exs or config/runtime.exs:

config :cscf, :cscf_hosts,
  pcscf: [
    {host: "10.4.12.165", port: 9060, label: "P-CSCF 1"}
  ],
  icscf: [
    {host: "10.4.12.166", port: 9060, label: "I-CSCF 1"}
  ],
  scscf: [
    {host: "10.4.12.167", port: 9060, label: "S-CSCF 1"}
  ]

The control panel provides navigation tabs for each CSCF component:

  • P-CSCF - /pcscf - Registration contacts and hash tables
  • I-CSCF - /icscf - S-CSCF list, NDS domains, sessions
  • S-CSCF - /scscf - Registrations, dialogs, IFC management
  • Diameter - /diameter - Diameter peer status and control
  • Logs - /logs - Live log viewing

P-CSCF Management​

URL: /pcscf

Features​

The P-CSCF panel displays registered contacts and hash table information from P-CSCF instances.

Registered Contacts Tab​

Displays all current IMS registrations visible to the P-CSCF:

ColumnDescription
IMSISubscriber IMSI or Contact identifier
StateRegistration state (registered, unregistered)
ExpiresTime until registration expires
PathSIP Path header for routing

Operations:

  • Click on row to expand and view detailed contact information including:
    • Full AoR (Address of Record)
    • UE IP address
    • Path details
    • Statistics (max slots, records)

Hash Tables Tab​

Manage P-CSCF hash tables. See Hash Table Operations below.

Real-time Updates​

The P-CSCF view automatically refreshes every 5 seconds to show current registration status.

I-CSCF Management​

URL: /icscf

Features​

The I-CSCF panel provides monitoring of I-CSCF operations including S-CSCF selection and session tracking.

S-CSCF List Tab​

Displays all configured S-CSCF servers known to the I-CSCF:

  • ID: S-CSCF identifier
  • Name: S-CSCF FQDN
  • Capabilities: Number of supported capabilities

NDS Domains Tab​

Shows trusted NDS (Network Domain Security) domains configured on the I-CSCF.

Sessions Tab​

Displays active I-CSCF sessions including:

  • Call-ID: SIP Call-ID
  • S-CSCF Candidates: List of S-CSCF servers considered for assignment
    • S-CSCF name
    • Selection score
    • Age (time since candidate was added)

Hash Tables Tab​

Manage I-CSCF hash tables. See Hash Table Operations below.

S-CSCF Management​

URL: /scscf

The S-CSCF panel is the most feature-rich, providing comprehensive registration, dialog, and IFC management.

List Registrations Tab​

Browse all active registrations with pagination:

Features:

  • Pagination controls: Offset and limit for large registration databases
  • Registration details for each IMPU:
    • Public user identity (IMPU)
    • Registration state
    • Slot number
    • Contact details with User-Agent and expiry
    • Call-ID

Quick Actions for each registration:

  • Lookup: View detailed IMPU information
  • Dump IFC: View Initial Filter Criteria for the user
  • Test IFC: Test IFC matching for simulated calls
  • Deregister: Administratively remove registration

User Location Tab​

Query and inspect user location data:

  • View raw user location status from S-CSCF
  • IMPU Lookup form: Query specific public user identity
  • Displays complete registration details including contacts, state, and metadata

Dialogs Tab​

Manage active call sessions (dialogs):

ColumnDescription
Dialog IDh_entry:h_id identifier
Call-IDSIP Call-ID
FromCalling party URI
ToCalled party URI
StateDialog state

Operations:

  • End Dialog: Terminate specific call (sends BYE)
  • End All: Terminate all active calls (with confirmation)

IFC Tab​

Initial Filter Criteria tools for service triggering management:

Dump IFC​

Retrieve and display all IFC rules for a given IMPU:

  • Public identity
  • Private identity
  • Service profile count
  • Filter Criteria for each service profile:
    • Priority (execution order)
    • Default handling (SESSION_CONTINUED vs SESSION_TERMINATED)
    • Application Server name
    • REGISTER inclusion flags
    • Trigger Point details:
      • Condition type (DNF or CNF)
      • Service Point Triggers (SPTs):
        • METHOD, HEADER, SESSION_CASE, REQUEST_URI, etc.
        • Negation flags

The IFC display includes:

  • Color-coded priority badges
  • Expandable trigger point logic explanations
  • DNF (Disjunctive Normal Form) = OR of ANDs
  • CNF (Conjunctive Normal Form) = AND of ORs

Test IFC​

Test which Application Servers would be triggered for a simulated session:

Input:

  • URI (subscriber public identity)
  • Direction (originating or terminating)
  • Method (INVITE, REGISTER, MESSAGE, SUBSCRIBE)
  • Request URI (destination)

Output:

  • Registration status
  • Matching IFC count
  • List of triggered Application Servers with IFC index

Hash Tables Tab​

Manage S-CSCF hash tables. See Hash Table Operations below.

Diameter Peer Management​

URL: /diameter

Features​

Monitor and control Diameter peer connections (Cx, Rx, Ro interfaces).

Summary Information​

Dashboard displays:

  • Realm: Diameter realm
  • Identity: Diameter Origin-Host
  • Peer Count: Number of configured peers
  • Workers: CDP worker count
  • Queue Length: Pending transactions
  • Connect Timeout: Connection timeout (seconds)
  • Transaction Timeout: Transaction timeout (seconds)
  • Accept Unknown Peers: Policy flag

Peer List​

Table of all Diameter peers:

ColumnDescription
FQDNPeer fully-qualified domain name
StateConnection state (I_Open, Closed, etc.)
StatusEnabled or Disabled
Last UsedTime since last transaction
ApplicationsNumber of supported Diameter applications

Operations:

  • Enable Peer: Activate disabled peer
  • Disable Peer: Deactivate peer (with confirmation)
  • Click on row: Expand to view supported applications

Application Mapping​

The control panel automatically maps Diameter Application IDs to 3GPP interface names:

  • Cx/Dx (16777216:10415) - IMS Subscription/Authorization
  • Sh/Dh (16777217:10415) - User Data Access
  • Rx (16777236:10415) - IMS Media Plane Control
  • Ro (16777238:10415/0) - Online Charging
  • Gx (16777224:10415) - Policy Control
  • S6a/S6d (16777251:10415) - LTE/EPC MME-HSS
  • And many more (see source: diameter_live.ex)

Real-time Updates​

Diameter peer status automatically refreshes every 5 seconds.

Hash Table Operations​

Overview​

CSCF components use in-memory hash tables for performance-critical data. The control panel provides visibility and management of these tables.

Available Hash Tables​

Tables vary by CSCF type. Common examples:

Hash TableCSCFPurpose
imei_msisdnP-CSCFEmergency callback mapping
service_routesP-CSCFCached service routes
authS-CSCFAuthentication vectors
VariousAllComponent-specific caching

Hash tables are in-memory data structures used for performance-critical operations.

Viewing Hash Tables​

Access: Navigate to any CSCF panel → Hash Tables tab

  1. View list of all hash tables with statistics:

    • Table name
    • Item count
    • Size

  2. Select table to view entries

  3. Sort by name, items, or size

Viewing Hash Table Contents​

Click on a table to inspect all entries:

  • Key: Hash table key
  • Value: Stored value
  • Actions: Delete button

Managing Hash Entries​

Delete Single Entry​

  1. Select hash table
  2. Locate the entry
  3. Click Delete button (trash icon)
  4. Confirm action

Result: Entry removed from hash table

Flush Entire Table​

  1. Select hash table
  2. Click Clear Table button
  3. WARNING: Confirms before clearing ALL entries
  4. Confirm action

Result: All entries removed from table

Caution: Flushing tables can cause temporary service disruption:

  • imei_msisdn flush: Emergency callbacks may fail until re-registration
  • auth flush: In-progress authentication challenges will fail
  • service_routes flush: Next request will route via I-CSCF discovery

Logs Viewing​

URL: /logs

Features​

View application logs in real-time from the control panel.

Features (implementation in ControlPanel dependency):

  • Live log streaming
  • Log level filtering
  • Search and filtering capabilities

Monitoring and Metrics​

Prometheus Integration​

OmniCall CSCF exposes Prometheus metrics for monitoring and alerting.

Metrics Endpoint:

http://<host>:9090/metrics

Each CSCF host (P-CSCF, I-CSCF, S-CSCF) exposes metrics on port 9090. Configure Prometheus to scrape all hosts for complete visibility.

For a complete reference of all P-CSCF, I-CSCF, and S-CSCF metrics see the Metrics Reference.

Available Metrics​

The following metrics are exposed by the OmniCall CSCF control panel application. For CSCF component metrics (SIP, Diameter, IMS, etc.), see the Metrics Reference.

VM Metrics​

  • vm_memory_total - Total Erlang VM memory (bytes)
  • vm_memory_processes_used - Memory used by processes (bytes)
  • vm_memory_binary - Binary memory (bytes)
  • vm_memory_ets - ETS table memory (bytes)
  • vm_total_run_queue_lengths_total - Total run queue length
  • vm_system_counts_process_count - Process count
  • vm_system_counts_atom_count - Atom count
  • vm_system_counts_port_count - Port count

Phoenix HTTP Metrics​

  • phoenix_endpoint_stop_duration - HTTP request duration (milliseconds)
  • phoenix_router_dispatch_stop_duration - Router dispatch duration (milliseconds)

LiveView Metrics​

  • phoenix_live_view_mount_stop_duration - LiveView mount duration (milliseconds)

CSCF Backend Integration Metrics​

  • cscf_backend_request_count - Backend RPC request count
    • Tags: host, command, result
  • cscf_backend_request_duration - Backend RPC duration (milliseconds)
    • Tags: host, command
  • cscf_backend_error_count - Backend RPC error count
    • Tags: host, error_type

Grafana Dashboards​

Current Status: Metrics are exposed via Prometheus endpoint. Pre-built Grafana dashboards are not currently included but can be created using the available metrics.

Recommended Dashboard Panels:

  • Backend RPC latency by command
  • Registration count trends
  • Dialog count trends
  • Backend error rates
  • Erlang VM memory usage
  • LiveView mount performance

Integration​

Configure Prometheus to scrape metrics from all CSCF hosts:

scrape_configs:
  - job_name: 'cscf_pcscf'
    static_configs:
      - targets: ['pcscf1.example.com:9090', 'pcscf2.example.com:9090']

  - job_name: 'cscf_icscf'
    static_configs:
      - targets: ['icscf1.example.com:9090', 'icscf2.example.com:9090']

  - job_name: 'cscf_scscf'
    static_configs:
      - targets: ['scscf1.example.com:9090', 'scscf2.example.com:9090']

Best Practices​

Operational Guidelines​

Monitoring:

  • Monitor Prometheus metrics for system health
  • Watch for backend RPC errors
  • Track Erlang VM memory growth

Hash Table Management:

  • Avoid flushing tables in production unless absolutely necessary
  • Monitor table size growth for potential memory issues
  • Use selective deletion instead of full table flush

Troubleshooting:

  • Use Live Logs for real-time debugging
  • Check Diameter peer status before investigating registration failures
  • Verify CSCF backend connectivity if control panel shows errors

Performance:

  • Control panel auto-refresh is 5 seconds by default
  • Large registration lists use pagination to avoid performance issues
  • Hash table operations are read-heavy; minimize write operations during peak hours