Rebuilding VICIdial conf files after changes

VICIdial stores your phones, carriers, and routing in MySQL, but Asterisk reads plain text config files on disk. Something has to bridge the two. That something is the rebuild conf files step: it reads the relevant rows out of the database and regenerates the on-disk Conf file set Asterisk loads. Change a server setting in the admin and nothing happens to live calls until those files are rebuilt and Asterisk picks them up.

Most operators meet this the first time they add a phone or a carrier, hit save, and notice the change does not register until a minute later. That minute is the rebuild cycle.

What gets regenerated

The rebuild writes the SIP and IAX peer definitions, the Dialplan contexts, and the trunk entries from your database rows. On a modern build that means the pjsip wizard config for your SIP peer entries, the IAX2 config for inter-server links, and the extensions that route calls. Add a webphone in the admin and the rebuild is what actually creates its SIP registrar entry on disk so the phone can register. Change a Server trunk and the rebuild rewrites the trunk's peer block.

When the rebuild fires

It is driven by a flag in the database. When you save a change that affects Asterisk config, VICIdial sets a rebuild flag on that server. The Keepalive process — the cron-driven supervisor that keeps the dialer healthy — checks that flag every minute, and when it sees it set, it regenerates the conf files and reloads Asterisk. You can also trigger it by hand from the admin if you do not want to wait for the next keepalive pass.

sequenceDiagram
  participant Admin
  participant DB as Database
  participant KA as Keepalive
  participant AST as Asterisk
  Admin->>DB: Save phone or carrier change
  DB->>DB: Set rebuild_conf_files flag
  KA->>DB: Check flag every minute
  DB-->>KA: Flag is set
  KA->>AST: Write conf files and reload
  AST-->>KA: Config active

Triggering it on purpose

Two safe ways. Set the rebuild flag from the server's admin page and let the next keepalive cycle pick it up within a minute. Or run the keepalive conf-regeneration step directly over SSH for an immediate rebuild. Both end with an Asterisk reload, not a restart, so calls in progress are not dropped. Avoid hand-editing the generated files: the next rebuild overwrites them, and your edit vanishes. Put the change in the database where the rebuild will keep it.

If a phone or carrier you added is not working, a missed rebuild is the first thing to check. Our post-install checklist includes verifying the conf files match the database, and the VICIdial install guide explains where these files live in a fresh build.

How it behaves on a managed box

On VICIfast the keepalive runs from the moment your Single tenant box comes up — we provision it in under 40 seconds with the rebuild loop already live — so admin changes flow to Asterisk without you touching a cron. You keep root SSH, so a manual rebuild is always available when you want a change to take effect this second instead of next minute.

Make the change in the admin, let the rebuild carry it to disk, and verify the peer registered. See what each plan includes on our pricing page and get a box where the conf rebuild is already wired and running.