How to turn on call recording in VICIdial

Recording in VICIdial is controlled per campaign, not globally. There is no single master switch that turns audio capture on for the whole system; instead each Campaign has its own recording level. That is good news because it means a sales line and a collections line can have completely different rules, but it also means a new campaign starts with whatever default it inherited. This guide walks through where the setting lives and how to confirm calls are actually being captured.

Where the recording setting lives

Open the campaign you want to change and look for the Campaign Recording menu. This is the field that decides whether Call recording happens at all, and if so, who is in charge of it. It offers four values: NEVER disables all recording on the client, ONDEMAND lets the agent start and stop recording as needed, ALLCALLS starts recording whenever a call is sent to an agent, and ALLFORCE does the same but gives the agent no option to stop it. ONDEMAND is the default, so a fresh campaign already lets agents record manually even if nobody touched the setting.

If you simply want every call captured without depending on the agent remembering to press a button, ALLCALLS is the common choice. Set the Campaign Recording menu to ALLCALLS, save the campaign, and from that point any call routed to an agent on this campaign begins recording on its own. Pick the level that matches your policy and move on; the rest is plumbing.

The recording extension and format

Next to the recording level is the Campaign Rec extension. This controls how long a single recording can run and which Codec the audio is stored in. Extension 8309 records in WAV for up to one hour and is the usual default; 8310 records in GSM, also up to one hour, which produces much smaller files. By default a recording stops after one hour, so if you run long calls and need the whole thing, the maximum has to be raised on the server side before the limit will move. For most floors the one-hour cap never gets hit, so the default is fine.

How recording gets decided on each call

flowchart TD
  A[Call sent to agent] --> B{Campaign Recording level}
  B -->|NEVER| C[No recording]
  B -->|ONDEMAND| D[Wait for agent button]
  D --> E[Agent records on demand]
  B -->|ALLCALLS| F[Start recording automatically]
  B -->|ALLFORCE| G[Start recording and lock it on]
  F --> H[Audio saved to file]
  G --> H

Once the level is set, every inbound or outbound call follows the same path. The dialer checks the campaign level when the call reaches the agent, then either does nothing, waits for a button, or starts capturing audio. The recording is written to a file named from a pattern you can customize, and the result lands in the recordings directory on the server where it can be played back later.

Confirming it works

Place a test call, complete the Disposition so the call finishes cleanly, and then look for the file. If you chose ONDEMAND, remember you have to press the record button during the test or nothing will be saved, which trips up a lot of first-time setups. If you chose ALLCALLS or ALLFORCE, the file should be there with no extra action. If you want to understand how the four levels relate to each other in detail, our explainer on campaign recording modes breaks each one down, and the broader call recording guide covers storage, filenames, and retention.

Getting recording right on day one saves a painful audit later. If you would rather have the dialer, the recording paths, and the storage already configured and supported, see our pricing and have a working setup in minutes instead of an afternoon.