Your customer calls you. Their phones are down. You log into their 3CX portal — and yes, the trunk has been deregistered for three hours. You missed the alert because there was no alert. The customer found out before you did, and that's the conversation you're now having.
Every MSP managing more than a few 3CX instances has had this exact conversation. The 3CX admin console is excellent for managing one server. It is not built for the operational reality of an MSP managing twenty. This guide covers what 3CX monitoring actually means for MSPs, what to monitor, and how to do it without drowning in dashboards.
What "3CX monitoring" actually means for an MSP
When 3CX talks about monitoring, they usually mean the built-in dashboards inside a single instance — call quality reports, trunk status, extension activity. Useful, but designed for the IT person at one company managing their own PBX.
When an MSP says monitoring, they mean something different: a fleet view across every customer's 3CX instance, proactive alerting that catches problems before customers do, and an audit trail of what changed and when. The goal isn't visibility into one PBX — it's operational control across many.
The rest of this guide is about the MSP definition.
What you should actually be monitoring
Instance availability
The basics. Is the PBX up? Is the web interface reachable? Is the SSL cert valid? Sounds obvious, but a surprising number of outages come down to expired certificates and Linux services that crashed and didn't restart. If your monitoring tool isn't checking these every minute, you don't have monitoring.
SIP trunk registration
Most 3CX outages start here. A trunk deregisters, calls fail, the customer doesn't notice until lunch. Per-trunk monitoring with instant alerting is non-negotiable. If you have ten customers and forty trunks across them, you need a single view that shows all forty and screams when one drops.
Backup status and age
This is the silent killer. 3CX backups can quietly fail or stop running, and you only find out when you actually need to restore. Monitoring backup success and age (when did the last successful backup complete?) prevents the worst conversation you can have with a customer.
Disk and memory utilization
3CX call recordings, voicemails, and logs eat disk. A full disk doesn't just stop recording — it can take down services. Visual thresholds with warn/critical alerting catch this before it bites.
License utilization and expiry
You should know which customers are approaching their simultaneous-call license limit beforethey hit it, so you can have a proactive upgrade conversation. You should also know which licenses are expiring in 30/60/90 days, because nothing is worse than a customer's PBX going to read-only mode the morning of an audit.
Queue and ring group configuration
Especially for customers with call centers. Monitoring queue member counts and ring group configuration catches misconfigurations (an agent removed by mistake, a ring group pointing nowhere) before customers do.
Extension changes
An audit trail of every add, remove, or modification. Useful for compliance, billing, and the inevitable "why is this user no longer working?" forensic exercise.
Service health and version
Which 3CX version is each instance on? Which need patching? Which are out of support? A fleet view of versions is what tells you where your security and upgrade backlog lives.
Automatic remediation, where possible
The best monitoring tools don't just tell you something broke — they try to fix it first. When a 3CX service stops responding, an automatic restart attempt resolves the majority of cases without human intervention. The alert only escalates to a person if the automatic recovery fails. This is the difference between getting paged six times a week and getting paged six times a year.
Why the 3CX Partner Portal isn't enough
The 3CX Partner Portal is genuinely useful as an inventory— it lists every instance tied to your Reseller ID. It is not monitoring. There's no alerting, no health scoring, no mobile push, no PSA integration, and no audit trail.
Most MSPs use the Partner Portal as a starting point — but it's not the operational tool. Think of it as the source of truth for what you manage; you still need something on top of it that actually watches.
For a detailed breakdown, see Sikurd vs 3CX Instance Manager.
The alerting layer — getting the right alert to the right person
Monitoring is only useful if alerts actually reach a human in time to act. For MSPs, that means three things most legacy tools get wrong:
Mobile push, not just email. A 2am trunk failure email lands in an inbox that nobody opens until 8am. Push notifications to iOS and Android wake the on-call tech.
Escalation policies.If the first responder doesn't acknowledge in 10 minutes, the alert escalates to the next person. Without this, your most senior tech ends up being your only responder.
PSA integration. The alert should auto-create a ticket in ConnectWise, Autotask, HaloPSA, or Syncro — with the right priority and customer attached. Anything that requires a human to manually open a ticket will get skipped on a busy day.
Suppression of planned maintenance is the fourth thing. If you're scheduled to reboot a server at 11pm, you don't want the resulting "instance down" alert paging the whole team.
How MSPs monetize 3CX monitoring
Monitoring isn't just a cost — it's the foundation of a recurring revenue line for your 3CX business. The pattern most successful 3CX MSPs follow:
Bundle it into your managed services contract. A flat monthly fee per customer that includes proactive monitoring, backups, adds/moves/changes, and support for critical events. Customers pay for the peace of mind; you deliver it remotely without rolling a truck.
Use brandable reports to justify the fee.A monthly uptime and activity report — branded with your logo — is what your customer's CFO sees and approves the renewal against. Without it, the fee feels invisible.
Surface upgrade conversations proactively. When you can see a customer approaching their license limit, you can have a planning conversation instead of a fire drill. That conversation usually ends in an upgrade and bigger margins.
Stop rolling trucks.Most issues that used to require an on-site visit can be diagnosed and resolved remotely with good monitoring data. The savings are real — and they're yours, not the customer's.
Tooling — what's available
There are a handful of tools built specifically for managing 3CX at MSP scale. We've covered them in detail in our guide to the best tools for managing multiple 3CX servers, but the short version:
- Sikurd — modern SaaS, mobile-first, transparent pricing, AI fleet intelligence
- PBXMonitor — long-established, deep automation, quote-based pricing
- IPTechView (ABP Tech) — distributor-aligned, US-focused
- 3CX Partner Portal — free inventory tool, not real monitoring
- PRTG / Zabbix — general-purpose, requires significant custom configuration
For a direct comparison of the two purpose-built SaaS options, see Sikurd vs PBXMonitor.
What to look for when choosing
- Purpose-built for 3CXGeneric tools require you to build the 3CX-specific intelligence yourself
- Multi-instance fleet viewOut of the box, not a bolt-on
- Mobile push alertsTo iOS and Android
- SIP trunk monitoringAt the per-trunk level
- Backup age and success trackingNot just instance up/down
- License expiry and utilization tracking
- PSA integrationFor ticket auto-creation
- Escalation policiesSo the right person gets paged
- Automatic remediationService restart on detected failure
- Transparent pricingIf you can't see the price, the price is whatever they think you can pay
- No agents, no firewall changesConnect via 3CX's API
- Audit logOf platform actions
- Time to first instance monitoredShould be minutes, not days