Test Your Backup Communications Before an Outage
Matthew Leo · Published October 23, 2026 · Guides
A backup internet connection can fail at the same time as the primary one. Both services may use the same fibre route, building power, router, identity provider or upstream carrier. Organizations that depend on connectivity need to test the whole communications path, not merely buy a second subscription.
The lesson became unusually visible on October 14, when Ukrainian officials said military Starshield terminals stopped working reliably while moving at high speed. Drone, reconnaissance and logistics missions were disrupted before Ukraine later reported that in-motion service had resumed. The reasons remained disputed, and Europe had no immediate drop-in alternative for thousands of proprietary terminals.
Canadian organizations face different threats, but the planning problem is familiar. The Canadian Centre for Cyber Security recommends alternate telecommunications services, separation between primary and backup providers, provider contingency plans and regular testing.
Start with the work that must continue
List the business functions that cannot wait for normal service to return. Examples may include dispatch, safety monitoring, payment authorization, access to medication records or contact with remote staff. Assign each function a recovery time: the longest outage it can tolerate before the consequences become unacceptable.
Then estimate its minimum communications requirement. A voice channel and small text messages may survive on a weak connection that cannot support video meetings, cloud desktops or large file transfers. The backup plan should prioritize essential traffic rather than promise normal performance.
Trace shared failure points
Write down every component used by both the primary and alternate path:
- building power, uninterruptible power supplies and generator fuel;
- last-mile cables, towers or satellite terminals;
- routers, firewalls and Wi-Fi access points;
- DNS, identity and multifactor-authentication services;
- cloud applications and data stores;
- the people, credentials and vendor contacts needed to activate service.
Two providers do not offer meaningful separation if both enter the building through the same conduit or depend on the same upstream carrier. Likewise, a satellite terminal is not a complete fallback if the organization cannot power it, position it, authenticate it or move essential work onto it.
Define a degraded mode
Decide in advance what stops during an outage. Pause automatic software downloads, personal streaming, cloud backups and other high-volume traffic. Keep capacity for emergency calls, operational messaging and the smallest applications required to deliver essential services.
Document a manual alternative for decisions that normally depend on live data. That could mean printed contact lists, offline maps, paper intake forms or a local copy of the day’s schedule. Protect those materials because they may contain personal or operationally sensitive information.
Test activation, not just availability
A useful exercise begins with the primary service treated as unavailable. The response team should retrieve the plan without using that connection, power the alternate equipment, authenticate, route essential traffic and notify staff through a second channel.
Measure several things:
- Activation time: how long it takes before the first essential service works.
- Usable capacity: what the alternate path can sustain under realistic load.
- Coverage: which rooms, vehicles or remote sites can connect.
- Dependencies: which credentials or cloud services still require the failed path.
- Battery endurance: how long equipment operates without building power.
- Return to normal: whether queued work and records reconcile safely after restoration.
The Cyber Centre says business-continuity plans should be documented, communicated and validated through tabletop exercises or live simulations. Its medium-impact control profile specifically recommends alternate telecommunications services that do not share a single point of failure and calls for periodic testing.
Ask providers for evidence
Obtain the provider’s contingency plan and ask when it was last tested. Clarify whether emergency policies can change capacity, geography, speed or permitted uses. Record support contacts that remain reachable when the customer portal or normal phone system is unavailable.
A contract cannot eliminate every outage, but it can establish notification expectations, restoration priorities and the information needed to make an operational decision.
Keep the plan reachable
Store an offline copy of the activation steps, network settings, vendor contacts and decision authority. Review it when staff, equipment or contracts change. A backup path that only one absent employee understands is another single point of failure.
Mapletechie’s analysis of government-owned satellite capacity explains why control can matter alongside coverage. For most organizations, the practical first step is smaller: prove that an alternate path works before the primary one disappears.
Sources
- Canadian Centre for Cyber Security: developing a business continuity plan
- Cyber Centre telecommunications-contingency controls
- Reuters analysis of Ukraine’s satellite-service dependency
Tags: business continuity, Starlink Mobile, Canadian telecom, Canada