Why a VICIdial cluster needs a standby database
Every dialer and agent screen in a VICIdial cluster runs through one database. A standby is the live copy that takes over when that machine fails.
You've spread your floor across several dialers so no single machine can take it down. But there's still one machine every dialer depends on, and if it stops, the whole floor stops with it.
That machine is the database. The standby database exists for the day it fails.
The one machine everything runs through
In a cluster, the database holds every Lead (a contact record you dial), every Campaign (a set of leads and dialing rules) and every call record. In our layout the same machine also serves the agent screens and reports.
The dialers lean on it all day. They run Asterisk, the phone engine that places and answers the calls, and they query the database many times a second. VICIdial hands each answered call to a free agent through the database, and agents' browsers load their screens from it.
Compare what happens when each kind of machine fails:
- A dialer fails. The calls on it drop, and the other dialers keep the floor working. That's why we never build a cluster with fewer than two.
- The database machine fails. With no standby, nothing can hand out calls or load a screen until that machine is repaired or rebuilt and the data put back.
One failed server stopping the whole floor is exactly the problem a cluster is meant to remove. Two dialers remove it on the calling side. The standby removes it for the database.
What a standby actually is
A standby database is a second machine that replicates the database as it changes: a live copy, not last night's. If the database machine fails, the standby is promoted and takes over. That handover is called Failover.
flowchart TD
D1[Dialer 1] --> DB[(Database)]
D2[Dialer 2] --> DB
DB -. replicates every change .-> SB[(Standby database)]
DB --> X[Database machine fails]
X --> PR[Standby promoted]
PR --> N[Dialers use the standby]We give it the same hardware as the database itself, a 16-core AMD Ryzen 9 7950X3D with 128 GB and mirrored NVMe disks. The moment it's promoted, it carries the whole floor's database load, so it can't be the smaller machine.
How long the switch takes on your cluster isn't something we'll promise in a blog post. Recovery times are written into your agreement.
Why it sits right next to the database
It's tempting to put the standby in another city, just in case. In a VICIdial cluster that backfires. Every machine sits in one datacenter on a private network, because the dialers query the database many times a second, and a round trip between Germany and Finland is over 20 ms. That's enough to stall VICIdial.
A far-away standby would look fine while it waits. The day it's promoted, every dialer would be talking to a database 20 ms away, and the floor you just rescued would stall. Spreading across regions is an Enterprise conversation, not a default layout.
A standby is not a backup
So the standby sits alongside your backups, not instead of them. A backup strategy for a VICIdial server covers the other half.
If you're weighing whether one more machine is worth it, what dialer downtime actually costs you is a good place to start the math. The standby is already included in every cluster hardware total we publish.
Frequently asked
- No. A standby copies every change as it happens, mistakes included, so you still need separate backups.
- Not in our layout. A round trip between Germany and Finland is over 20 ms, enough to stall VICIdial once the standby is promoted.
- The calls on that dialer drop, and the other dialers keep the floor working.
› Is a standby database the same as a backup?
› Can the standby sit in a different datacenter?
› What happens if a dialer fails instead?
For the rest of the layout, read our guide to VICIdial clusters or what each part of a cluster does. Still small enough for one box? Pick a plan and keep it simple. When you're ready for a cluster, book a sizing call and see today's prices.
Related from VICIfast
- VICIdial cluster sizing, with real numbersA database, dialers and a standby, sized from load we measured on a live floor.
- A whole machine for your dialerEvery core yours, mirrored NVMe, delivered in minutes to days. Priced from live stock.
- The branded agent webphoneA WebRTC phone inside the agent screen on every server, in your colours.
- Dedicated-CPU plansAMD EPYC dedicated cores for steady volume or visible CPU steal.
About VICIfast
We run VICIdial servers for call centres, so you don’t have to. You get the dialer set up, secured and kept running — we handle the server, the updates and the backups. Bring your own phone carrier at no markup, or use VICIfast Voice, our own US calling and numbers, from $0.0085 a minute. From $49 a month per server.
Questions? Call +1 636 556 0022 and talk to someone who runs dialers.
Citing this article
VICIfast Engineering. “Why a VICIdial cluster needs a standby database”. VICIfast LLC, November 20, 2026. Retrieved from https://vicifast.com/blog/vicidial-cluster-standby-database
Have questions?
Related posts

Guides & tutorials
Running the VICIdial database on its own server
Guides & tutorials
We pushed a VICIdial database to 7 times a live floor: what slowed first

Guides & tutorials
What is a VICIdial cluster? The parts in plain words

Guides & tutorials
VICIdial clusters: how they work, when you need one, and how to size one
You might be interested in
Guides & tutorials
We pushed a VICIdial database to 7 times a live floor: what slowed first
Guides & tutorials
What hardware a VICIdial database server really needs
Guides & tutorials
What 166 live calls cost a VICIdial server: one measured shift

Guides & tutorials
How many dialer servers does a VICIdial cluster need?
VICIfast newsletter
Liked this? Get the next one in your inbox.
We ship the kind of stuff you just read - concrete, numbers-first, no drip. One email when a new post goes live. Unsubscribe in one click.
Comments
No comments yet - be the first.