Emergency equipment replacement is the work of finding, approving, buying, shipping, and licensing hardware after a failure with no local spare. Recovery time objectives assume the constraint is technical, but most of the elapsed time between failure and restoration goes to sourcing and physically moving the replacement. The engineering work is often the shortest phase.
Organizations test failover. Very few test procurement and logistics under emergency conditions, which is why the first real test tends to reveal that the approval chain, the shipping path, and the license transfer process were never designed to run at 3 a.m.
What Does Your Support Contract Actually Cover?
Often less than assumed, because service level agreements are widely misread. A four-hour response contract is not a four-hour restoration contract.
Confirm, in writing, for each critical platform:
- Whether the SLA covers response, part arrival, or restored service
- Which parts are covered and which are considered consumable or customer-replaceable
- Where the nearest stocking depot is and what it actually holds
- Whether the clock runs continuously or only during business hours
- What the contract says when the required part is unavailable
Coverage also expires. Equipment past end-of-service-life may still be running critical workloads while carrying no entitlement at all. Reconcile the support inventory against the actual production inventory — the two drift apart steadily, especially after migrations and consolidations.
Which Spares Strategy Fits Each Platform?
There are three practical positions, and most environments need a mix:
On-site spares — fastest possible recovery, capital tied up, and a maintenance burden. Justified for single points of failure and for platforms with long procurement lead times.
Vendor or partner depot stock — hours rather than minutes, no capital cost, dependent on depot inventory being what the contract implies. Verify what is actually stocked locally rather than trusting the coverage map.
Procure on demand — lowest cost, longest and least predictable timeline. Appropriate for equipment with genuine redundancy and tolerance for extended degradation.
Whichever position applies, record the decision per platform. A spares strategy that exists only as an assumption is discovered to be absent at the worst moment.
Sourcing Under Pressure
When there is no spare, the sourcing path depends on the age of the equipment.
- Current-generation hardware is generally available from the OEM or an authorized reseller, though configured systems can carry multi-week build times
- End-of-life hardware often exists only in the secondary market, where quality varies enormously and verification matters
- Some components — certain optics, memory configurations, controller boards — have unpredictable availability regardless of platform age
Identify secondary-market suppliers and establish relationships before an emergency. Buying critical hardware from an unvetted source during an outage means accepting unknown provenance, unknown condition, and no recourse, the counterfeit and tampering risk that NIST SP 800-161 addresses.
How Fast Can a Replacement Part Be Shipped?
Anywhere from a few hours by dedicated courier to overnight by expedited freight. Once a part is sourced, physical movement is a solvable problem if the options are known in advance.
- Dedicated ground courier — regional, direct, no terminal handling, typically hours
- Next-flight-out air freight — cross-country in hours rather than days, with a counter-to-counter handoff
- Chartered transport — expensive, but available when nothing else meets the window
- Standard expedited freight — cheapest of the expedited options, but still overnight at best and subject to terminal delays
Know which providers serve your sites, what their cutoff times are, and how to engage them outside business hours. A part sitting at an airport because nobody was authorized to arrange the final leg is a common and entirely avoidable delay.
The Access Problem
Equipment arriving at 2 a.m. is useless if it cannot get through the door.
- Confirm after-hours dock access and whether a facility escort is required
- Establish how a courier who is not on the access list gets admitted
- Verify colocation providers' remote-hands availability and engagement procedure
- Keep current after-hours contacts for facilities, security, and the provider
- Know the delivery constraints — dock height, freight elevator hours, building access rules
Colocation facilities in particular enforce access lists strictly, and the person who can add a name to that list at 2 a.m. should be identified before you need them.
Why Does Licensing Stop So Many Recoveries?
Many enterprise platforms bind licenses to a chassis serial number or hardware identifier. Replacing the hardware invalidates the license, and rehosting requires a vendor transaction that may only run during business hours.
- Document the license rehost process for each critical platform, with the specific portal or contact
- Maintain current license files and entitlement records outside the affected system
- Confirm whether emergency rehost is available outside business hours, and how it is invoked
- Keep saved configurations and golden images stored somewhere the outage cannot reach
A replacement chassis that arrives on time and cannot be licensed until Monday morning has not solved the problem.
Pre-Authorize the Decisions
Emergency procurement fails on approval chains as often as on availability.
- Set a pre-approved spending threshold for emergency hardware, and name who can invoke it
- Put master agreements in place with suppliers so a purchase order is not being negotiated during an outage
- Establish a payment method that works outside business hours
- Document all of it in the incident runbook rather than in someone's memory
Asking for a signature at 3 a.m. from an executive who does not answer their phone is not a procurement process.
Test the Path
Run the exercise the way a failover test is run under a NIST SP 800-34 contingency plan. Pick a critical platform, walk the full sequence, and time it:
- Identify the failed component and confirm entitlement
- Locate the replacement — depot, spare, or market
- Execute the approval and purchase steps
- Arrange transport and confirm site access
- Confirm the license and configuration restoration path
Do this on paper first, then with live phone calls where possible. The result is usually a realistic replacement timeline that differs substantially from the one assumed in the recovery plan — which is the entire point of running it before it matters.