What the Internal Process Logs page shows you

The Internal Process Logs page shows you a seven-day history of every back-end VICIdial process across your entire cluster — how many times each one was launched, when it was launched, and how long it ran — so you can see exactly what broke and when it broke.

Where to find this page

Internal Process Logs lives in the same Admin Utilities section as the Server Versions page — the level-9 link at the bottom of the main reports menu. No date picker, no filters. The page loads all servers in your cluster and shows the last seven days of data automatically.

Which processes appear in the logs

VICIdial runs its core functionality through a set of long-lived Perl scripts managed by a screen keepalive system. These are the processes the Internal Process Logs tracks:

• AST_* processes — the Asterisk-adjacent scripts that handle call generation, answering, and routing. These are the heart of the outbound engine.
• VDAD (VICIdial Auto Dial) — the core auto-dialing process that feeds calls into the Hopper and out to the Carrier. If this process is down, no outbound calls go out.
• listen — the process that monitors agent browser connections and keeps the real-time display in sync.
• fastlog — the high-throughput log writer that records call events and agent activity to the database. A disrupted fastlog causes data gaps in reports.

Each of these runs inside a screen session and is watched by the keepalive script. If a process exits for any reason — a crash, a memory fault, a failed database query — the keepalive restarts it automatically. The Internal Process Logs captures each restart event.

The three data points per process

For each process, the page shows:

• Launch count — how many times the process was started in the seven-day window. A healthy process that has been running since the last planned restart shows a count of one or a very small number.
• Launch time — the date and time each instance was started. This timestamps when a fault occurred: if the launch time is mid-campaign on a Tuesday afternoon, something went wrong at that moment.
• Run duration — how long each instance has been running. A process launched seven days ago and still running has a seven-day duration; a process launched five minutes ago and already gone has an effectively zero duration.

sequenceDiagram
  participant KA as Keepalive script
  participant P as Backend process
  participant DB as VICIdial DB
  participant L as Internal Process Logs
  KA->>P: Launch process
  DB->>L: Write launch record with timestamp
  P->>DB: Running normally
  P->>P: Crash or exit
  DB->>L: Write run duration at exit
  KA->>P: Detect exit and relaunch
  DB->>L: Write new launch record

How this data helps you

The combination of launch count, launch time, and run duration turns this page into a fault timeline. When a manager reports that calls were not going out for 20 minutes last Thursday, you open Internal Process Logs and find the VDAD process launched three times in that window, each instance running for only a minute or two. That pattern tells you the process was crashing and being restarted repeatedly, and the timestamps tell you exactly when it started and when it stabilised. Without this page you would be guessing from server-level logs.

For the companion view of current server state — load, channels, disk, and active status — read the overview of the Server Performance Report, which shows the resource-level picture that often accompanies a process crash.

Internal Process Logs is one piece of a complete monitoring habit. For the full framework, see the guide to monitoring VICIdial server health and capacity.

Want a managed VICIdial box where process-level logging is configured and accessible from day one? Spin up a managed VICIdial box on VICIfast and have it running in under 40 seconds.